Authors
Frax Core Team
Summary
Remove all swaps for canFRAX and canFXS from any bridge. Deprecate support for any wrapped representation of FRAX+FXS on any other chain leaving only natively issued FRAX+FXS on Ethereum mainnet.
Background
Frax Finance pioneered the “canonical stablecoin” token architecture released in 2021 for canonical FRAX and canonical FXS. This allows each chain such as Moonbeam, Fantom, Polygon etc to have a native canFRAX ERC20 token contract which both the protocol and supported bridges can swap 1 to 1 between (up to certain limits).
For example, the canFRAX ERC20 contract on Polygon allows multiple different bridges to create canFRAX stablecoins by depositing 1 to 1 wrapped FRAX tokens representing bridged FRAX from Ethereum mainnet. If a user bridges 100 FRAX from Multichain to Polygon, they can deposit the 100 Multichain wrapped FRAX token to swap for 100 canFRAX on Polygon. If another user bridges 500 FRAX through the official Polygon bridge, they too can also bring their 500 wrapped FRAX and swap it for 500 canFRAX as well (minus swap fees). All AMMs, lending protocols, and AMOs use canFRAX of their respective chain so liquidity is not fragmented and 1 single address for FRAX becomes canonical per chain. This has very powerful network effects to make 1 FRAX token the default stablecoin rather than multiple kinds of bridged FRAX each with low liquidity and use cases.
MakerDAO recently announced they are releasing a similar feature (oddly named the exact same as canFRAX called canDAI): Canonical DAI Stablecoin Upgrade Aims to Help Move DAI Off Chain | by Zero Mass | BanklessDAO | Jul, 2022 | Medium This proves that the canonical stablecoin architecture is the best way for a stablecoin to grow on other chains and build a network effect around a single token address.
However, there is one specific problem with the canonical stablecoin system: the bridge tokens that are swapped 1 to 1 for canFRAX are held within the protocol’s ownership and thus expose the risk of multiple bridges. Additionally, a single bridge being hacked removes liquidity for canFRAX users since some of them won’t be able to swap back to Ethereum (the native home for the Frax protocol and all FRAX, FXS, FPI, FPIS tokens) due to part of the liquidity being worth $0 . We saw this occur with the Harmony 1Bridge hack where some canFRAX on Harmony is now stranded without the ability to bridge that amount of canFRAX back to Ethereum. This is also currently happening with canFRAX on Moonbeam as we speak due to the Nomad bridge hack. Essentially, the current structure of canFRAX is that it is not technically a native liability of the Frax protocol but simply a pooled representation of of all bridged versions of FRAX on the destination chain that diversifies the bridge risk but does not remove it. This is why when a bridge is hacked like Nomad or Harmony, users of canFRAX are rightfully concerned what happens to their canFRAX stablecoins on their chain since the backing of those canFRAX has significantly taken a hit.
This governance proposal aims to change the above structure and remove bridge swaps between canFRAX+canFXS stablecoins. This means that if you hold canFRAX on any chain, it is fully backed by the Frax protocol’s value and collateral on Ethereum mainnet and the CR is correctly displayed for all canFRAX and Ethereum mainnet FRAX combined. However, in order for you to be able to bridge canFRAX and canFXS back to Ethereum, there will need to be liquidity providers in Curve pools, Fraxswap, other AMMs that pair wrapped bridge tokens such as anyFRAX with canFRAX creating swap pools. This means the risk of bridge hacks and bridge risk is entirely offloaded to LPs of these bridge pools that pair canFRAX with different wrapped versions of FRAX bridged from Ethereum. For example, in order to exit your canFRAX on Fantom through the Multichain bridge to Ethereum, there must be a Curve pool that has liquidity between anyFRAX-canFRAX so that a user swaps for anyFRAX, then uses that to return to Ethereum. Bridge partners can apply for FXS gauge pools for these “bridge liquidity pools” but the protocol will no longer support them with POL or 1 to 1 swappers for canFRAX or canFXS.
Count-intuitively, should this proposal pass, it could make canFRAX one of the safest assets on non-ETH chains because it has the full faith and commitment to redeem against FRAX on Ethereum L1 while having 0 bridge risk. Chains that use a specific bridge as the predominant way to bring all tokens to their chain such as Fantom using Multichain bridged assets or Moonbeam using Nomad assets means that if those bridges are hacked, almost all tokens are worthless on their entire network except for canFRAX or canFXS. Hypothetically if Multichain was hacked, then USDC, USDT, ETH, WBTC, and almost all other tokens would be worth 0 on Fantom since all liquidity pools and projects use the wrapped bridge version of those tokens through Multichain. Fantom has almost no natively issued blue chip tokens other than FTM itself.
In the above example, canFRAX would still retain its full peg and value regardless of any other bridge risk or what any chain uses as their main bridge since canFRAX holders would not be at all exposed to any kind of bridged asset whatsoever. Put in another way, there could be situations where every asset on another chain is worthless other than canFRAX and their L1 token if this governance proposal passes.
Proposal
Remove all swaps from canFRAX and canFXS on all chains with their supported bridges. Remove all protocol owned liquidity and all protocol ownership of wrapped bridge FRAX & bridged FXS on every chain.
For: Deprecate all 1 to 1 swaps between canFRAX & canFXS in their respective contracts on all chains. Remove all protocol owned wrapped/bridged FRAX+FXS from balance sheet.
Against: Do nothing