[Company Watch] Float Protocol, $Eth Backed Soft-Peg Stablecoin

There are a lot of new projects launched recently, and some are aiming at a new generation of stablecoin. Despite a short history, stablecoins have had a few round of evolutions — some failed, and more are coming.

Earlier we have covered briefly on Fei Protocol, without going into the thinking behind the design. This article is a quick summary of Float, another Ether backed stablecoin similar to Fei, but different in many aspects. We will not go into depth in this article, but we will write another more lengthy and academic article on the design of stablecons, a preview of which can be found on our twitter.

https://floatprotocol.com/#/pools is a launching, in a couple of weeks time, a new stablecoin FLOAT, backed by a basket of coins (although all Ether in version 1). There are a few features:

• It’s funded by Ether, from the minting ceremony to later stabilisation process, Ether is the major payment for FLOAT
• It’s soft-pegged, meaning it’s not pegged to $1 or a fixed fraction of $eth, but somewhere in between. For the time being, we call it soft-pegged.

Details are all the official docs and as an appendix of this article, we have listed a number of explanation paragraphs by founders of Float in discord. Below are some salient points:

Value of FLOAT

First of all, ΣFLOAT minted=ΣETH in Basket. This means the total value of FLOAT in circulation should be equal to the ETH raised. You can think of this as the case of Hong Kong, the government issues $HKD, which is pegged to $USD to a large extent.

Secondly, as $eth has price movements and are volatile. The designer of Float does not wish FLOAT to be as volatile. So there are two variables to be added to the above equation. The below is given in the official docs:

But we tweak it a bit:

ΣETH in Basket = ΣFLOAT minted X Basket Factor X Target Price

If target price is given, a change in $eth price will change the Basket Factor. E.g. If $eth drops 10%, the Basket Factor changes from 100% to 90%. The protocol should maintain the Basket Factor as 100%, unless it’s changed by the governance vote (interesting — if people choose not to have a 100% back stablecoin — like the parliament votes to switch from gold-backed to legal tender).

The target price is a computed number, based on historical prices of $eth, e.g. TWAP of the last few days or hours. And it’s constantly changing. Details on how Target Price will change is not revealed by the team yet.

How would FLOAT target price move when $eth price changes?

The team has done a simulation of Float’s Target Price in 2020. Vault factor is Basket Factor. From the graph, Float is not always at the original Target Price, but is much more stable than $eth.

You can argue that ETH-USDT pair in Uniswap also behaves the same way. But well, Uni-V2 tokens are not designed for spending.

The Stabillization Mechanism

When FLOAT is over the Target Price, the protocol launched an auction for arbitraguers to buy newly minted FLOAT at a discount to Target Price.

When FLOAT is below the Target Price, the protocol launched an auction for arbitraguers to sell their FLOAT to the protocol at a premium to the market price.

Both auction methods are Dutch, so it creates competitions for arbitraguers to act fast. Both auctions provide immediate arbitraging opportunities, so arbitrageurs make profits within the same block. Arbitraguers do not take risks of holding anything, like a bond in the case of other algo stablecoins (e.g. Empty Set Dollar or Mith Cash), and wait for the price to recover.

The Platform Token

The protocol has a token, BANK, which is designed as governance token, and also as a cushion token when the Basket Factor is too low. I.e. when $eth drops too much, and the Target Price revision falls to catch up, then the protocol will run short of $eth to pay for the Float in auctions. The the Protocol mints BANK to pay for a portion of the FLOAT buying.

In other cases, users or the protocol buy BANKs and burn them to make sure it’s in balance.

The Overall Design Consideration

Float Protocol is an interesting and not aggressive design. It has a lot of flexibility in managing FLOAT. Ideally, the protocol seeks to have a stablecoin as stable as possible. But the worst case is only that FLOAT will be reduced to a hard peg of $eth. In future, the basket composition can be more than $eth, e.g. BTC, DAI, or anything else.

If you think about it — this is how the modern central bank of a small economy works, think about Hong Kong or Singapore. Independent currency; as stable as possible; a sufficient foreign reserve of a basket of other countries currencies like USD, CNY, EUR; central bank has the power and market tools to be directly involved in the money market to control the exchange rate.

You might argue that Float Protocol’s design is not a perfect solution. For instance, arbitraguers always take away some value from the system, and after all, FLOAT holders and BANK holders are bearing this loss of value. It’s true and there’s no direct cure for that.

There’s always a cost of maintaining a currency system — this is the seigniorage cost. No matter how smart a design is, a currency system itself does not generate value. If a currency system manifests without generating value, it’s called inflation. And this is what defi strives to change.

Appendix: Notes by the founders of Float Protocol in Discord:

George H | Float — 03/24/2021

So this is actually a slightly out of date chart (it’s called the Basket Factor now — we thought this was clearer). Purple is the change in Float’s daily target price. Blue is the Basket Factor

[9:38 AM]so, as you can see, the target price tracks the basket factor quite nicely, but smooths out the volatility

Paul M | Float — Today at 6:02 AM

It does. If FLOAT, BANK and ETH all get sold at once and price drops, then it’s tough for any protocol to withstand. That being said we’ve got a couple of tricks: 1) If this is a sustained drop (not a Black Thursday style event) then the FLOAT price finds a new normalised target price so that it’s well supported — 100% — and we’re back to the standard market case. This is a trick that’s uniquely enabled by being a floating currencies, and also make sense — float should decrease in price as the underlying market decreases. 2) If FLOAT drops quicker that ETH, then we can actually resupply the basket using that difference. 3) Because FLOAT unlike stablecoins, finds a new normal BANK will gain value again after the crash. That means that its value is significantly less likely to decrease — you can actually profit from a crash with BANK.

George H | Float — 03/24/2021

So we know FLOAT ‘floats’ (changes in a value every day depending on what’s happening to the crypto market and demand for FLOAT)

[9:58 AM]but the protocol always has a ‘target price’ in mind (this is the thing that actually changes daily)

John L | Float — 03/22/2021

Target price moves based on overall market sentiment over a period of time. If the demand for FLOAT is high and the basket is well supported the target price increases. If the demand for FLOAT is low and the basket isn’t well supported the target price decreases. The precise details will be released in our interactive litepaper which is coming soon, where you will be able to play with the different scenarios

(Serenity Team, 30 March 2021, Twitter: https://twitter.com/SerenityFund)