FIP2 and the New Bond System

Written by @Sovandar

At the time of this post, over 2 million FSD bonds have been purchased.

It’s been quite a month for FSD. The bootstrap phase at the start of January created too many tokens too quickly; the result was oversupply, effectively “hyperinflation” and a sharp price crash to below 1 cent.

Obviously this is less than ideal — however, the core principle of algorithmic stablecoins is that when price is trading above the $1 peg, new tokens are minted to increase the supply and reduce the price; and below the peg, incentives are provided to new buyers and holders of the token, to help return to peg. With a substantial oversupply, this system was strained — but has not broken.

The development team have spent the time since the end of bootstrap hard at work, and the first major change to the protocol — FIP2 — is now live, replacing the “coupon” system used in other algorithmic stablecoins, with a new “bond” system. The exact details of this amendment are covered in the previous article Announcing FSD Bond System: A Completely Novel Approach, but an overview of the system, the thinking behind it, and a broad idea of how it works, are the purpose of this article.

Other algorithmic stablecoins derived from Empty Set Dollar (ESD) all inherited a “coupon” system that activates when price is trading below the $1 peg; this allows users to burn their tokens for coupons, at a specific “premium”, so that they can end up with more tokens than they originally burned (e.g. with a 40% premium, every burned token can earn 1.4 coupons, which can become 1.4 tokens at a later date).

The aim of a coupon is to give people a reward in return for permanently burning some or all of their tokens, reducing supply and thus mitigating the sell pressure keeping price below the peg. When price does return to the peg, coupons become “redeemable”, meaning they can be swapped out for more tokens than were originally burned — but if price does not return to peg within the 30-day lifespan of the coupon, it expires worthless, and the burned tokens are simply gone.

For those who know trading terminology, a coupon is effectively a form of binary option, a high-risk trading method; for those less familiar, a coupon is like placing a bet on a roulette wheel — if you win, it’s great, but if you don’t you’ve lost all the money you bet with.

ESD and the next largest fork, Dynamic Set Dollar (DSD), have both had severe problems with their coupon systems. After a prolonged expansion event in ESD that inflated its token supply to over half a billion, they found that most people decided not to ‘bet’ with the coupon system at all — leaving price stranded well below peg, and with participation dropping. They have had to move to de-risk their coupons to try to encourage new interest, but they still expire after 30 days and re-mint new tokens, which just worsens the oversupply problem again. Meanwhile, DSD have found that despite huge participation in coupons, even minor changes in governance have become very controversial since DSD participants with coupons cannot vote, even though governance changes often affect them (including by increasing the risk of their coupons expiring) — forcing emergency proposals to de-risk their coupon system and replace it entirely.

While the basic idea behind coupon expiry sounds reasonable — if there is too big an oversupply, tokens should burn permanently, supply will reduce, and buyers can raise the price more easily — it is far from that simple. The example of ESD shows that if trust in the protocol’s rapid return to peg is lost, people don’t want to buy coupons and risk losing their money — so there is no participation, and price might never return to peg under those conditions.

The example of DSD though, shows that when participation is high, a large coupon expiry event simply wipes out the protocol’s biggest supporters and most active participants, effectively punishing the very people whose participation is most important for the project to continue — to put it mildly, that’s not a good way to build trust in a new and novel project!

Even ignoring the major issues that ESD/DSD have highlighted, there is a bit of an elephant in the room regarding coupons as well; they have proven very effective at pushing prices of these tokens away from their lows, but have not proven effective at pushing price higher once it’s “near” peg but still below it — they tend to get stuck above 70 cents each, for long periods of time, when nobody wants to buy more coupons. The reasons are much debated, but a major one is that when price is near peg, new buyers are taking the same risk with their coupons, but for a proportionately lower reward. Consider investing $1,000 in coupons when the token is 50 cents, and earning a 40% premium — you would buy 2,000 tokens and get 800 tokens premium, a profit of 180% if the coupon redeems. Buying at 99c, you would get 1,010 tokens plus 404 tokens premium, a profit of only 41% — set against the possibility of losing the entire $1,000 investment, that’s not a lot of incentive.

FSD’s solution is to replace coupons entirely, in favour of a “bond” system. There are some key differences to coupon systems, the major one being that bonds will never expire; so there is no risk of your funds just vanishing when a deadline is hit. This means that the risk participants take is simply the time spent waiting, not that they can lose all of their funds if they time it wrong. In effect, FSD’s approach is that rewards should be to encourage people to lock up their tokens to help reach peg — not punitively burn their tokens if it takes “too long”. With the bond system people only have to wait, not correctly guess when price will reach the peg again.

The other major change is that bonds don’t all become redeemable at the same time; they have a required target price that needs to be reached first. The closer the bond was to the peg when it was bought, the closer its redemption price is to $1 (but still above peg) — so buyers at 95c still have incentives to buy bonds, without worrying that they are just locking themselves in at the back of a long queue. However — because price is nearer peg for these bonds, they have a lower premium to reflect the easier redemption criteria.

The highest redemption price for a bond is $1.20 per token — which is a long way below the “maximum expansion” cap for FSD; it is highly unlikely that price would return to $1 and never get to $1.20!

The intent of all of these changes to create the bond system, is to make it much safer for people to participate when price is below the peg; and, because of that, send a clearer signal to new buyers that existing participants are confident in the price returning to the $1 peg. New buyers still have incentives to buy bonds even at prices close to peg, and the phasing effect of the target redemption prices should help keep FSD tracking its peg more closely in the long term, than other algorithmic stablecoins have managed to date.

With the new bond system live for barely 48 hours, already a quarter of all FSD has been put into bonds — over 2 million. Some large wallets are showing interest, actively participating in the bond system. All are signs of FSD recovering from the bootstrap supply inflation, with an active team supporting novel changes and developing something unique in the algorithmic stablecoin space.

It’s early days yet — but FSD is heading in the right direction, and we encourage people to participate in the new bond system; the more participants, the faster we can return to peg!