The case for canonical asset bridging: Secure value transfer and cross-chain interoperability
Tldr; Bridge solutions can use canonical assets or representative assets to transfer value. While using representative assets has some benefits, canonical asset bridging is a more secure approach because it does not increase trust assumptions.
Key takeaways:
- Bridges use canonical assets or representative assets to transfer value.
- While using representative assets has some benefits, it adds security risks.
- Using canonical assets is more secure because it does not add trust assumptions.
- Across is cheaper and faster than bridges that use representative assets, and it can offer these benefits without making security trade-offs.
Crypto may be down from its peak, but its multi-chain era is in full swing. The total value locked in Layer 2 has soared since 2021, to say nothing of the myriad new chains hitting the market this year. In recent months, Ethereum alternatives have begun migrating to Ethereum Layer 2s, and so have centralized exchanges. Outside of Ethereum, other smart contract chains are increasingly gaining popularity.
In this new world, users need robust bridge solutions to move their assets between networks. Bridges take different approaches to help users achieve their goal and they often make trade-offs. Native bridges, which are built and used by Layer 2 chains, use lock and minting to produce canonical assets. Third-party bridges mint representative assets, which creates a new validation mechanism. In all third-party bridges, introducing a new validation mechanism introduces security risks. There are other ways bridges use canonical assets. As using canonical assets does not increase trust assumptions, it is the only secure way to transfer value between chains.
Across’ design differs from others in the cross-chain ecosystem. Like others, it uses canonical assets, which minimizes security risks. Crucially, it also adopts an intents-based framework. This model is the optimal approach to bridging as it offers user experience benefits without adding security risk. We detail how below.
How bridges transfer value
In today’s cross-chain ecosystem, transferring value is the dominant use case for bridging. Most bridges use lock and mint mechanisms to transfer value. There are several different types of lock and mint bridge.
Canonical bridges (also known as native bridges) use lock and minting to help users transfer canonical assets between chains. They do not increase trust assumptions because the user does not have to trust a new contract; they only need to trust the security of the chain they are interacting with. The minting relies on the same validation methods that the chain as a whole relies on. The Optimism Bridge is an example of a canonical bridge.
Stablecoin bridges use lock and minting to help users transfer canonical stablecoins between chains. They are similar to canonical bridges in that the user does not have to trust a new contract; they trust the stablecoin and its issuer. Circle’s CCTP is an example of a stablecoin bridge.
Third-party bridges use lock and minting to help users transfer representative assets between chains. This adds a security trade-off because the user has to trust the asset they receive at the origin. When the user mints the representative asset, they’re trusting the bridge that they’ll be able to redeem it for a canonical asset in the future. They also trust the consensus of the chain they’re interacting with. Wormhole is an example of a third-party bridge.
As canonical asset bridging does not increase trust assumptions, it is the only secure way to transfer value.
Alternatively, bridges can use delivery vs. payment or intents mechanisms to transfer value. Rather than creating representative assets, they act as liquidity layers on top of canonical assets.
Delivery-vs-payment (DvP) bridges use liquidity pools to help users transfer canonical assets between chains. With DvP designs, the user deposits their funds, their deposit gets verified, and then they receive the canonical assets at the destination. This approach does not increase trust assumptions, but it has other drawbacks. Stargate is an example of a DvP bridge.
Intents bridges use intents to help users transfer canonical assets between chains. With intents, the user expresses their intent, a market maker (i.e. a relayer) fronts the order, and the user receives the canonical assets at the destination. As with canonical bridges and stablecoin bridges, this approach does not increase trust assumptions. Across is an example of an intents bridge.
Why using intents and canonical assets wins
Canonical bridges offer security benefits over other designs because they do not increase trust assumptions. However, they suffer from user experience drawbacks as transfers are slow.
Stablecoin bridges can be thought of as a subsect of canonical bridges, where the user only has to trust the stablecoin issuer. But they also suffer from limitations as they are limited to only one type of token.
Third-party bridges have some user experience advantages as they don’t require external liquidity and can fill orders faster than canonical bridges. However, as they use representative assets, they increase trust assumptions. The user has to trust the asset they receive and gets burdened with the security risk.
In third-party lock-and-mint bridges, no one pays for security. Users simply pay gas costs to lock up their assets in exchange for representative assets. This creates a risk burden.
Delivery-vs-payment bridges do not need to use representative assets, but they have other disadvantages. Verifying orders has high gas costs and users must wait for finality on two chains. This negatively impacts the user experience.
Intents bridges maintain security while offering an optimal user experience. As relayers front orders with canonical assets, users do not take on any additional risk. Using intents can also offer a faster, cheaper way to move between chains.
We believe Across’ approach is the winning bridge design. By using intents, Across can offer an optimal user experience without compromising on security.
The cost of using intents to bridge canonical assets
We have determined that using representative assets is a sub-optimal approach to bridging because no one pays for security.
The alternative is to use canonical assets. This approach requires external liquidity to fill orders quickly. The bridge can only function if LPs commit capital to liquidity pools.
Across uses canonical assets through an intents-based framework. When users indicate that they want to transfer their assets, relayers compete to fulfill their requests at the destination. There is a trade-off here, as the relayers expect a return on their capital.
But the cost of this loan is extremely low.
Let’s assume relayers are willing to lend out their capital at a 10% interest rate for a duration of one hour (this figure is an approximation based on market rates).
The 10% rate is annualized, and there are 8,760 hours in one year. That means the cost of the loan is just 0.001142%, or roughly 0.1 basis points. We can use the below formula to calculate the cost of a relayer loan:
Annualized loan rate ÷ (number of days in a year x number of hours in a day) = Cost of a one-hour loan
10% ÷ (365 x 24) = 0.001%
We can apply this formula to any sum to calculate the loan cost.
If the relayer lends out $1,000 for an hour, the cost would be ~$0.01.
If the relayer lends out $100,000 for an hour, the cost would be ~$1.
If the relayer lends out $10 million for an hour, the cost would be ~$100.
In addition to the costs to pay for relayer loans, LPs take a fee for committing their capital. Across’ intents-based architecture also enables gas optimizations such as bundled relayer repayments to reduce costs and protocol level rebalancing, which improves capital efficiency.
We can use the following formula to determine Across’ cost of using canonical assets:
Short-term relayer loan + LP fees — gas optimizations
As relayers only make short-term loans and the system makes gas optimizations, the costs of using canonical assets is very low. So while there is a cost to building on top of the security of canonical assets, the costs are lower than the gas fees to bridge representative assets through a third-party bridge. Across can offer this saving without making security trade-offs. Users benefit from low costs; they do not have to take on risk or pay for gas-intensive verification.
Canonical assets and the cross-chain future
While we’ve entered crypto’s cross-chain era, the ecosystem is still young. We’ve seen several early experimentations in cross-chain interoperability, and we can expect to see more in the future.
Soon, there will be many more chains and demand for bridging solutions will increase. To date, bridges have catered to market demands with representative assets, but this approach is not secure.
While there are costs associated with using canonical assets, using intents can offset them. With this in mind, we can draw two clear conclusions.
Canonical asset bridging is the only secure way to transfer value, and intents designs are the best way to bridge canonical assets.
Thanks to its intents-based framework, Across is the cross-chain ecosystem’s fastest, cheapest bridge solution. It is also the winning case study for canonical asset bridging. We’re confident that this approach will help crypto evolve into a secure, cross-chain ecosystem.
This piece was compiled with contributions and editing from Ryan Carman, Across Product Lead and Hart Lambur, UMA co-founder.
