Fixing Stablecoin Designs
Why We Started Kokoa Finance
Why Cryptoasset-backed Collateral?
There are two types of decentralized stablecoin. One is the collateral-backed model, and the other is the algorithmic model. The former has been proven to be very robust even during the bear market. The longstanding success of DAI(MakerDAO’s stablecoin) is a testament. Since the value of the collateral-backed stablecoin is always backed by collateral assets, its holders rarely worry about their stablecoin being insolvent or a bank-run induced by some financial panic in the market. As long as the underlying assets are sound, the risk can be adequately managed. On the other hand, the robustness of the algorithmic models is quite confusing in the sense that such models use circular reasoning to back their stability.
Algo… Does it Work?
A Brief Look at the Dual-Token Algorithmic Stablecoins
Even though Kokoa didn’t adopt the algorithmic model, it is worth reviewing other models. The dual-Token algorithmic model consists of a native stablecoin, whose value is guaranteed by its governance token, and a governance token whose utility comes from rewards generated by the seignorage and transaction fee of its stablecoin.
This is how this model works — when the price of the stablecoin rises, more stablecoins are minted to buy back the governance token; when the price falls, more governance tokens are minted to buy back the stablecoins. We should note that the seignorage profit generated during expansion periods should actually be considered as debt because the governance token guarantees the price peg. Considering that, the cash flow generated from transaction fees is the most crucial factor that explains the theoretical intrinsic value of the governance token.
In short, this is how such systems work — the inherent value of the governance token is justified by the expected transaction fee revenue it generates.
Risks of Dual-Token Algorithmic Stablecoins
Here is the logical flow of such model: (i) the stablecoin is stable, (ii) so people prefer it as a means of payment, (iii) and such payments generate transaction fees, (iv) the governance token accrues an intrinsic value from the generated fees, (v) and the stablecoin is stable because the governance token guarantees it.
Do you see the problem now? The whole logic is based on the assumption that the stablecoin is stable somehow, or at least many people believe so.
When everyone has such faith, this becomes a self-fulfilling prophecy. Because everyone believes in the stability of stablecoin, every step of the logic makes sense so that the governance token can justify its intrinsic value. Unfortunately, however, this flywheel can also rotate the other way around. People may start to doubt the price stability due to market conditions, FUDs, and so on. And, such doubts can also become another self-fulfilling prophecy.
Suppose the price of the stablecoin becomes volatile. As the advantage it has as a medium of exchange diminishes, the expected future cash flow generated from transaction fees should be adjusted downwards. Because the expected cash flow is changed, so should the theoretical value of the governance token. As a result, the system’s price guarantee is shaken.
To not fall into this vicious cycle, the following must hold. First, the market cap of the governance token should be at least worth more than its stablecoin. And, the transaction fee revenue it generates should be large enough to justify (i) the market cap of its governance token, and (ii) the risk premium for the potential dilution under deflationary periods.
Of course, there could be a successful Dual-Token model; yet maintaining the ideal equilibrium is not an easy task. There are many internal and external variables to take into account to manage this algo-stablecoin ecosystem seamlessly.
Tri-Token Algorithmic Stablecoins.
Another well-known model is the Basis-like algorithmic stablecoin model, which adopts a tri-token model consisting of a share token, a bond token, and a stablecoin. For the sake of explanation, let’s call them tShare, tBond, and tUSD! The stability mechanism of such models is explained as follows.
Similar to the dual-token model, when the price of tUSD is higher than $1, the system mints more tUSD out of thin air and distributes them to tShare holders as dividends. As the supply of tUSD increases, it restores its price to its peg. Conversely, when the price of tUSD is below $1, a different upward adjustment mechanism takes place. Unlike the dual-token model, the tri-token model introduces another kind of token called a ‘bond token,’ which is named ‘tBond’ in our example.
The detailed mechanism behind tBond may vary, but the overall ideas are the same. During the deflationary period, tBond tokens are minted and sold in tUSD, usually at some discount from its face value. tBond is a bond-like financial product that promises its holders that the system will repurchase it at its face value in tUSD in the future when the price of tUSD is equal to or higher than $1. For instance, when the tUSD price is at $0.9, the system may mint 100 tBond, its face value is 100 tUSD, and sell it to users at 90 tUSD.
Suppose John bought this 100 tBond with 90 tUSD. Ten days later, the price of tUSD recovers back to $1. Then the system will mint 100 tUSD and buy John’s 100 tUSD with it. As a result, John successfully earns 10 tUSD in 10 days. Pretty sweet!
Risks of Tri-Token Algorithmic Stablecoins
Obviously, this model bears some problems too. The logic behind its stability mechanism has circularity. People buy tBond under the deflationary period because of the belief that the tUSD will recover its peg. But why? It could be the belief that the current demand shortage is just a short-term event while the long-term demand for tUSD will recover. Another possible answer is the belief that demand shortage is not a short-term event, but there are enough fools in the market nevertheless, who believe so to purchase tBond with their tUSD leading to absorbing the supply surplus (or enough fools that think there are enough number of such fools… yadi yada). If the former condition always holds — which is impossible — tBond does not need to exist in the first place. So, let’s focus on the latter for argument’s sake.
During a deflationary trend, buying tBond is something that you want to avoid because (i) it has no explicit duration; (ii) its upside is limited to its nominal tUSD value; and (iii) you have no idea how long it would take for tUSD to recover its long-term demand. If tUSD recovers its long-term demand if ever, it means that you have to wait indefinitely for the redemption, or you might even risk losing all of your investment. Imagine the central bank of Zimbabwe issuing a ‘zero-coupon perpetual bond’ at an 80% discount that you can redeem when 1 ZWB recovers its initial value — would you buy the bond?
Because of the risks above, a rational market will demand a relatively higher risk premium on such bonds. That premium must be significantly higher than the risk-free rates of tUSD. Considering that stablecoins usually have a quite high risk-free rate, the tUSD rates must be pretty high too. Therefore, the system must issue a larger nominal value of tBond compared to the actual scale of the demand deficit.
Even if the market absorbs this newly issued tBond, the total amount of liabilities that the system bears is not reduced at all. In fact, the debt just shape-shifts itself and charges an extra cost. When the price of tUSD recovers, the system will have to issue new tUSD tokens to pay for the face value of the tBonds it issued.
Here’s the catch — the amount of newly issued tUSD is always much larger than the initial demand shortage that the bonds were issued to cover in the first place. To put it simply, the system has issued 150 tUSD amount of tBond to cover a demand shortage of 100 tUSD. After the gap has been closed, it has to issue another 150 tUSD. This leads to a demand shortage of 150 tUSD. Indeed, the initial problem has exacerbated during the process.
So why does this happen? The answer lies in the incentive design. During the inflationary period — where everything is going well — the governance token holders take all the juicy fruits called the ‘seigniorage.’ On rainy days, however, they are not the ones that take the hit. The ones who take the hit first are the stablecoin holders, and some of them should voluntarily give up their liquidity in the hope that someday they will be compensated.
To sum up, as long as these issues are not handled properly, such models cannot reliably guarantee stability. If a stablecoin system cannot ensure its stability, it is indeed meaningless to discuss the demand for its stablecoin, not to mention its monetary policies.
Kokoa’s Choice
A Few Pitfalls of the Existing Crypto Collateral-backed Models
Due to the risks mentioned above, Kokoa Finance adopts a collateral-backed model to offer more robust and secure stablecoin.
However, this model has a weakness — capital efficiency. For example, 15B+ USD worth of collateral assets are locked in MakerDAO to mint and maintain 6B USD worth of DAI stablecoin. Those 15B+ USD worth of tokens are just locked in the CDP contract and are not utilized in any manner. This is indeed a massive waste of capital. The opportunity cost makes maintaining the DAI circulation in the market itself very expensive.
If minting stablecoin is too expensive, the system cannot effectively supply enough stablecoin for the DeFi ecosystem that utilizes it. It does not even require taking Economics 101 to know that an adequate amount of currency should be provided for any economy to grow.
Kokoa Finance utilizes the collateral assets to generate low-risk yield on Klaytn’s DeFi ecosystem to address the very problem. Furthermore, Kokoa Finance is planning to add additional features that can maximize the multiplier effect on KSD while not compromising the soundness of the monetary system.
Related Articles
Website: https://kokoa.finance
Telegram (KR): t.me/KokoaFinance
Telegram (Global) t.me/KokoaFinanceGlobal
Twitter: https://twitter.com/KokoaFinance
Medium: https://medium.com/kokoa-finance
