Since Lightning Labs’ unveiling of the Taro protocol, stablecoins issued directly on the Bitcoin blockchain have once again become a hot topic. However, this is nothing new. The first stablecoin, Tether, was released on the Bitcoin blockchain through the Mastercoin (now known as Omni) protocol, which allowed for the issuance of additional tokens on the Bitcoin network. Originally introduced on the Bitcoin network, stablecoins have now moved to other blockchains as a result of the network’s limitations (such as the block size limit and the fee event in 2017). Starting with Ethereum, the trend has spread to other, less expensive and more controlled blockchains. No matter how decentralized the blockchain you issue them on, the value of a centrally issued stablecoin comes from the fact that it can be redeemed by a single centralized authority that can refuse to redeem it. The stablecoins themselves are not decentralized by placing their issuance on a blockchain, hence this practice serves no purpose beyond facilitating interoperability with native blockchain-based assets.
As far as censorship resistance goes, there is no point in processing stablecoin transactions on the Bitcoin blockchain, thus I think it’s excellent that we’ve moved on to other blockchains. If the issuer believes that the coins were used in illegal activity or were stolen, they are within their rights to refuse redemption of those coins. Stablecoins gain no actual censorship resistance from being issued and transacted on Bitcoin; the sole benefit is a modest reduction in the complexity of processes like atomic swaps for Bitcoin.
On the other hand, it changes the incentives behind Bitcoin. With the upcoming merging and shift to proof-of-stake on the Ethereum network, debates have arisen over the impact of stablecoins on the network’s consensus layer. Circle, the USDC’s issuer, has stated that they will only validate USDC transactions and honor redemptions on the PoS network. After the merge, any requests to redeem USDC on the other Ethereum forks would be ignored and not honored. This makes perfect sense as USDC is a reserve-backed stablecoin tied to dollars held in reserve by Circle. Only one set of stablecoins issued on a network can be redeemed at any given time, hence it would be absurd and impossible to respect redemptions on both sides of any fork. The reserve dollars do not suddenly multiply by two once that network splits in two, as the USDC tokens do.
This dynamic, however, grants stablecoin issuers disproportionate sway over the consensus of the network on which they have issued their coins. Ethereum’s usability and transaction volume are greatly bolstered by USDC. Following the merging and fork, all Ethereum users who wish to continue using their UDSC will be forced to transfer to the winning chain, regardless of their personal feelings on Proof-of-Work vs. Proof-of-Stake systems, or the split itself and which chain they choose to use. In order to spend their USDC, users must engage with the PoS chain. Tokens are needed to cover transaction costs while using USDC, therefore there is a kind of forced demand for them.
The same dynamic will occur with stablecoins based on Bitcoin. The issuers of stablecoins on the Bitcoin blockchain would have the same clout as Bitcoin miners during Bitcoin forks if Taro or the revival of the original Omni Tether token caused widespread issuance and transaction of stablecoins on the Bitcoin blockchain. If Bitcoin is widely used as a platform for stablecoin issuance and usage, then this would increase the demand for Bitcoin (since it will be required to pay transaction fees) and the income of Bitcoin miners (since they will be able to sell their Bitcoins for a profit). The asset’s demand and the mining industry’s ability to make money are then at the mercy of the stablecoin issuer.
All of the demand and miner profits after a fork go to the fork that the issuer chooses to support for redemptions. This can happen during a chainsplit, hard fork, or soft fork if the issuer deems a feature is undesirable and actively works to block its activation. The more the role stablecoins play in driving demand for the asset and blockspace, the greater their impact in such a scenario. If stablecoins make up 10% of a miner’s income, and the issuer of those coins takes a different path than the community at large during a fork, then those miners will need to move 10% of their hash rate to the new chain in order to keep making money. A 40% change would necessitate a 40% adjustment in hashrate.
To the same extent, Lightning Node operators gain from routing fees. Revenue from dollar payments routed through the network could dry up on the side of a fork that stablecoin issuers do not honor redemptions for if a significant amount of network activity is driven by users exchanging BTC for stablecoins at the edges. To get the revenue from stablecoin use, those node operators will need to run and operate nodes on the other fork.
Since stablecoins are so widely utilized on Ethereum’s network, Bitcoin is not miraculously immune to the problems Ethereum is experiencing just because it lacks Ethereum’s complex and insecure scripting language or Bitcoin’s daily use of on-chain decentralized exchanges. In this respect, the problems Ethereum is experiencing are entirely attributable to economic incentives, and they apply just as much to the Bitcoin network.
Bitcoin users should carefully consider whether it is in their best interest to promote and use such systems built directly atop Bitcoin, and whether the risks of doing so are worth it in light of how they interact with the incentives of the network. There are other blockchains that can run quasi-centralized networks, and there are even systems like Elements (on which Liquid is built) that can do this. To make an atomic exchange is not difficult at all. The technologies exist to develop systems for stablecoins that can host them externally to the Bitcoin network and allow for easy interaction with it.
Since atomic swaps on a single blockchain are marginally easier than atomic swaps between two blockchains, should we really introduce a big new centrally controlled variable to the incentives of the entire network? Although I can only speak for myself, I would not agree.