The operation mode of NFT royalties: design, challenge, and new ideas

Author: a16zcrypto

Translation: Gate.io (ERC6551 Space)

The royalty automatically executed in secondary sales has always been an important value proposition of Non-fungible Tokens. In an ideal world, creators can set royalties on-chain, and the royalties will be automatically paid whenever their works are sold anywhere on the Internet, without relying on the market or other third parties to fulfill the royalties out of goodwill.

However, the royalty of Non-fungible Tokens has never been enforced on-chain, which has been misunderstood. The demand for on-chain enforcement of royalties exceeds the progress made in achieving this goal. The challenge lies in distinguishing which Non-fungible Token transfers should involve royalty payments, and which transfers are of other types, such as self-transfers between the user’s own Wallet, or gifting Non-fungible Tokens, etc.

The newer royalty designs attempt to address this challenge by identifying different types of transfers and enforcing royalties at the appropriate times—but there is a significant trade-off between strict royalty enforcement (ensuring royalty payments) and composability (the ability for Non-fungible Tokens to interact with other on-chain applications).

Therefore, in this article, we discuss the advantages and disadvantages of existing NFT royalty designs, as well as the balance between royalty enforcement and composability. Then, we introduce two new methods of NFT royalty that utilize incentive mechanisms to drive market participants to respect royalties. Our goal is not to advocate for a specific approach, but to help builders consider different NFT royalty designs and their associated trade-offs.

First of all, what is “composability”?

Modularity is one of the core features of Open Source software, allowing developers to combine, modify, and mix project snippets like building blocks without permission to create interesting new applications.

Applications can combine with Non-fungible Tokens in two basic ways - read (check ownership) or write (facilitate transfer):

Reading (checking ownership) means verifying Block chain data. Applications can combine Non-fungible Tokens by verifying ownership as a ‘gate’ for further operations. For example, Non-fungible Token owners can gain the right to claim another Non-fungible Token, play games, participate in governance processes, obtain licenses to use Non-fungible Token media content, or attend conferences or concerts. In addition, people can use Non-fungible Tokens to associate on-chain data with their Wallet Address (for example, a Non-fungible Token containing a username, allowing the person who owns the Non-fungible Token to use that username on social media).

**Writing (facilitating transfer) indicates updating the blockchain state.**Transfer of Non-fungible Token will update on-chain who owns the Non-fungible Token. In the simplest example, people can directly transfer the Non-fungible Token to another Wallet. Applications can also combine this transfer feature in one of the following ways: (1) representing the owner to transfer the Non-fungible Token (e.g., in the case of the Non-fungible Token market) or (2) temporarily hosting the Non-fungible Token (e.g., custody of off-chain transactions, lending protocol for Non-fungible Token rental, or accepting Non-fungible Token as collateral for lending protocol).

It is important to distinguish between the different types of Non-fungible Tokens in terms of their composability. When we refer to ‘composability’ in this article, we mainly mean the composability of ‘writing’ or ‘transferring’.

Although anyone can verify the ownership of Non-fungible Token on-chain, the existing royalty design limits which Wallets and Smart Contracts can execute transfers or initially own Non-fungible Tokens. Restrictions on ‘writing’ may stifle the opportunities for using Non-fungible Tokens in Decentralized Finance, games, sharing ownership through Multisig, or even gifting to friends, as well as applications where Non-fungible Tokens own other Non-fungible Tokens.

Now let’s break down the existing royalty solution and weigh the pros and cons in more detail.

Existing solutions: blacklist and Allowlist

One key reason for the difficulty in enforcing royalties on Non-fungible Tokens (NFTs) is the challenge of distinguishing between transfers that are sales - and should therefore incur royalties - and transfers of other types. Specifically, due to the default implementation of the NFT standard, NFT smart contracts cannot determine whether a transfer is related to a sale or not. Existing solutions attempt to provide more context about transfers on-chain by restricting transfers (i.e., whether the transfer is a sale or occurred through a specific marketplace).

The most popular design for enforcing NFT royalties—blacklists and Allowlists—adopts different approaches to restrict transfers, and limits the composability of “write” or “transfer” along with it.

Both designs revolve around preventing transfer on two levels:

1. Prevent the transfer facilitated by markets or applications that bypass royalty payments.

2. Prevent transfer to certain account types: externally owned account (EOA) (the majority of Wallets used today) and Smart Contract account. In other words, there are certain restrictions on which types of accounts can own Non-fungible Tokens.

Therefore, regardless of the design used by the creators, they face a significant trade-off, depending on how their Non-fungible Token smart contract implements the transfer ‘prevention’: the stricter the transfer prevention by the creators, the lower the composability of the Non-fungible Token.

Blacklist

A blacklist is a specific Smart Contract Address or application list that is not allowed to facilitate the transfer of Non-fungible Tokens. Creators will add the Address of specific markets or applications that do not pay royalties to the blacklist within their Non-fungible Token Smart Contract; if a Non-fungible Token holder tries to transfer their Non-fungible Token through a blocked application, the transaction will fail. You can find more information about the blacklist here.

You can think of them as firewalls on your computer: you can freely browse the web, but the firewall will block sites it deems unsafe. Here, the “firewall” will block known applications that do not respect copyright.

Advantages:

By default, Non-fungible Tokens can freely combine with most applications. This is because the blacklist takes an optimistic view that most applications will fulfill royalties.

Creators can immediately protect royalties. Creators can close it by adding contracts that bypass royalties to the blacklist.

Cons:

Bad actors can bypass the blacklist. Bad actors can always launch a new market that bypasses copyright fees and is not on the blacklist.

The blacklist cannot proactively prevent the circumvention of royalties, only passively prevent it. New markets can be launched at any time. Creators are forced to play a game of cat and mouse, monitoring which markets are bypassing royalties and adding them to the blacklist.

The last point is the biggest challenge. In order to make the blacklist effective, the creator needs to constantly monitor new on-chain applications, track every possible new Smart Contract market, analyze it, and then decide whether to block it. This is a difficult job; even existing markets may need to be reviewed again as they upgrade Smart Contracts.

Excluding an application that bypasses copyright from the blacklist means missing out on payment. In addition, there is a “leaky bucket” problem: if even a market that bypasses copyright is not blocked, there may be disproportionate transaction flow to that market.

A potential solution is to entrust the management of the blacklist to a third party. However, this reintroduces a reliance on intermediary institutions to help enforce royalties, giving this entity market power, and may lead to various other consequences beyond the scope of this article.

Allowlist

The Allowlist explicitly specifies the Smart Contract Address or application that is uniquely allowed to facilitate the transfer of Non-fungible Tokens. Through this strategy, creators only allow markets or applications that guarantee royalty enforcement. Non-fungible Token holders can only transfer their Non-fungible Tokens through the Smart Contract on the Allowlist; if they attempt to transfer Non-fungible Tokens using a market not on the Allowlist, the transfer transaction will fail.

The existing Allowlist design also includes some optional components, such as: (1) restricting which types of Wallets are allowed to own Non-fungible Tokens, usually only allowing external accounts (EOA) rather than Smart Contract accounts; (2) restricting whether peer-to-peer (p2p) transfers are allowed.

Advantages:

Non-fungible Token transfers cannot be made through applications that are not on the Allowlist, such as marketplaces that bypass royalties. The Allowlist only approves transfers facilitated by Smart Contracts that creators know will fulfill royalties. By default, all other marketplaces are blocked. Depending on the implementation of the Allowlist and the marketplace, it can be difficult to transfer Non-fungible Tokens through marketplaces that bypass royalties (see the following disadvantages).

Creators do not need to track and add new markets that bypass royalties as in blacklist mode. After creators add one or more markets that fulfill royalties to the Allowlist, the urgency of monitoring new applications is lower or long.

Cons:

Creators need to approve individual applications that facilitate the transfer of Non-fungible Tokens. Whether it is a blacklist or an Allowlist, some level of on-chain monitoring is required. For blacklists, creators need to monitor applications that bypass royalties to avoid missing out on royalty payments. The Allowlist imposes permissioned composability. Creators won’t miss out on royalty payments, but they may miss out on innovative new applications built around Non-fungible Tokens. Let’s assume a developer builds a unique Non-fungible Token marketplace concept (also enforcing royalties!). The developer needs to contact Non-fungible Token creators, prove that they fulfill royalties, and request to be added to the Allowlist of each Non-fungible Token. This is a high-friction process.

It is still possible to bypass royalties, depending on the implementation of the market and the restrictions on the transfer of Non-fungible Tokens by creators. For example, if Non-fungible Tokens are allowed to be sold for $0, it is still possible to bypass royalty payments through a market on the Allowlist. In this case, someone can build a market on the Allowlist that bypasses royalties and facilitates $0 sales on the royalty-compliant market while transferring the actual payment on the side. Since the sale price is $0, the royalties for the creators are also $0 (i.e., 5% of $0 is $0).

The Allowlist may be too strict. The strictest version of the Allowlist also restricts the types of Wallets (EOA or Smart Contract accounts) and peer-to-peer (p2p) transfers that are allowed to have Non-fungible Tokens. Restricting Smart Contracts from owning Non-fungible Tokens is intended to prevent Non-fungible Token wrapping (see below), but in a world where everyone uses Smart Contract Wallets, this may be too strict. Restricting p2p transfers means that any transfer must go through a market on the Allowlist. The reason for this restriction is to prevent creators from missing royalties in over-the-counter (OTC) p2p sales. Restricting p2p transfers makes it difficult for Non-fungible Token holders to transfer Non-fungible Tokens between their own Wallets or directly between friends.

Consideration

The allowlist and blacklist introduce a balance between strict copyright enforcement and open composability. The blacklist model defaults to open composability, but it is easier to bypass royalties. Using the allowlist makes it easier to enforce royalties, but significantly limits the applications with which non-fungible tokens can interact.

This trade-off is not just about the comparison between blacklists and allowlists: any way of restricting the applications and methods through which non-fungible tokens can interact will limit the composable and functional nature of non-fungible tokens.

Improved technical methods may reduce the degree of this trade-off. But the fundamental problem still exists.

Exploring a New Framework for NFT Royalties

Creators are still testing the Allowlist, but as more and more long Non-fungible Token use cases emerge, it is worth exploring the boundaries beyond the blacklist/Allowlist model to improve the balance between royalty enforcement and composability.

The strategy we are exploring here redefines the problem and existing royalty mechanisms from the perspective of incentive design: our goal is to introduce incentives to drive the Non-fungible Token market and/or encourage consumers to actively choose to respect royalties. This provides the possibility of greater long-term composability in principle.

We demonstrate two different approaches below. The first mechanism improves upon the Allowlist model, making it more open, more composable, and more encouraging of permissionless innovation based on Non-fungible Tokens. The second mechanism, which we call “recapture rights,” provides strong incentives for consumers to use royalty-respecting markets when selling Non-fungible Tokens, allowing for significant royalty payments while maintaining open composability.

Our goal is not to propose a single ‘solution’, but to expand the range of choices: how to ensure that creators receive longer royalties without limiting composability and without relying entirely on good faith?

Method 1: The mechanism that combines the Allowlist and stake mechanisms

We can expand the existing Allowlist model through the stake mechanism, allowing the market and other applications to obtain Allowlist membership without permission.

Currently, creators must manually add markets or applications to their Allowlist, and third-party developers must request permission from the creators to be added. This slows down innovation and adoption of new applications, and shifts the responsibility of verifying that new applications perform royalties onto the creators. Delegating the management of the Allowlist to third parties may also slow down this process.

Introducing the stake model to obtain Allowlist membership will allow new applications to commit royalties through staking funds or other resources as a guarantee (‘optimistically’ trusting and then verifying, rather than assuming malicious behavior). By default, Non-fungible Token holders can immediately interact with new applications that provide appropriate stake; if the application behaves improperly, the creator can reduce the stake and remove it from the Allowlist. We can even imagine a hybrid model where, if the application proves its honesty over time, the creator can formally add it to the Allowlist and refund the stake.

There are some unresolved issues with this design approach. We list them here so that others can share further thoughts and research.

How do creators implement arbitration for reducing stake? The standard for reducing stake - i.e. whether royalties have been executed - may be challenging to detect and prove on-chain. Application developers need to trust that creators will not reduce their stake and remove them from the Allowlist when they should not be reduced.

Who should receive the slashed stake? On the one hand, giving the slashed stake to the creators may be a way to partially compensate them for the royalties they lost due to triggering the slashing event. But if the slashed stake does not belong to the creators, they would have no incentive to maliciously slash. Inspiration can be found from the EIP-1559 transaction fee mechanism of Ethereum, where the base fee of the transaction is burned instead of being sent to validators.

Should the size of the stake be less long? The value of the stake needs to have a certain relationship with the royalty amount that the application may generate for a given creator. Smaller stake amounts may be suitable for unpopular or niche applications. However, markets that facilitate a large number of Non-fungible Token sales may need to provide longer stakes, and the level of stake may need to gradually increase with the increase in the value of the collection and volume.

Do we need to stake among long Non-fungible Tokens? If so, how should we proceed? Developers may need to stake resources for each individual Non-fungible Token they want to aggregate, which is a significant burden. However, if developers stake a collection and prove its honesty, this may drop the burden for other Non-fungible Token creators to add new applications to their Allowlist. Similarly, we can imagine a strategy where the market uses a large amount of single stake to commit royalties in a wide range of collections.

Method 2: Mechanism Allowing ‘Right to Recycle’

The right of recycling is a new approach that encourages royalty payment for each Non-fungible Token sale through incentive measures, thereby transcending the trade-off between enforceability and composability (as well as blacklisting/Allowlisting). The core of this strategy is the improvement of the definition of on-chain ‘ownership’ of Non-fungible Tokens.

Each NFT can have two potentially different record owners, which we call asset owner and ownership owner:

The asset owner is the Wallet holding the Non-fungible Token (commonly referred to as “owner” today);

The owner of the ownership is the Wallet that pays the royalties (or ownership transfer fee, see below) to the Non-fungible Token creator.

This design method helps ensure that creators can obtain longer royalties without limiting composability.

Recovery Mechanism

If the asset owner and the ownership owner of the Non-fungible Token are different, that is, if the asset owner’s Wallet is different from the ownership owner’s Wallet, then the ownership owner can reclaim the Non-fungible Token to their Wallet at any time. The asset owner can become the ownership owner by paying the ownership transfer fee to the creator, thereby eliminating this ‘recovery risk’.

The right of recovery is not a lease, but it has similarities with leasing Non-fungible Tokens. For example, ERC-4907 is a standard for ‘Leasing Non-fungible Tokens’, which also has the concept of ‘owners’.

To simplify, we assume that the only way to transfer ownership is by paying the ownership transfer fee. However, in practice, there may be other ownership transfer mechanisms, such as automatically transferring ownership after a long enough time, or designing a mechanism for creators to directly trigger the transfer of ownership to the current asset owner.

In this model, the ownership transfer fee becomes the new “royalty”; a market that respects royalties will bundle the payment of the ownership transfer fee in the sales transaction. Note that this means that royalties no longer directly correlate with the sale price; the ownership transfer fee is a fixed cost, in contrast to the “percentage of sale price” fee historically used for NFT royalties. However, creators can choose to update the ownership transfer fee over time.

The risk of the owner reclaiming the Non-fungible Token helps to distinguish through people’s behavior which transfers of Non-fungible Tokens are sales (subject to royalties) and which are not. Specifically, this new ownership model incentivizes the payment of royalties for Non-fungible Token transfers between buyers and sellers, as otherwise the seller can immediately reclaim the Non-fungible Token after ‘selling’ it and receiving payment.

At the same time, this framework allows for free transfer between personal wallets or transfer as gifts.

Let’s take a look at how this works in practice through a few transfer examples:

1. If I transfer the Non-fungible Token to my own personal Wallet**, only the ownership of the asset will be transferred to the new Wallet, and my original Wallet will still be the owner of the ownership, but I don’t have the risk of recovery from myself.

2. If I transfer Non-fungible Token as a gift to fren**, only asset ownership will be transferred, and I will retain ownership. My fren can use it at will (including selling it; we will discuss how the market should handle this below), and can rely on social trust that I will not revoke Non-fungible Token. If my fren wants full ownership, they can pay the ownership transfer fee to the creator at any time. Alternatively, I can pay the ownership transfer fee when sending the gift Non-fungible Token.

3. If I sell through the market or trade over-the-counter outside the market (e.g., you give me 100 USDC, and I directly transfer Non-fungible Token to you)**, the buyer is strongly encouraged to pay the ownership transfer fee to eliminate the risk of me reclaiming the Non-fungible Token after receiving the buyer’s money.

Does the market need to change its way of working to adapt to this model?

In principle, it is not necessary. However, the right of recovery means that any Non-fungible Tokens purchased on the market have the risk of being recovered, which is a bad user experience - the buyer’s Non-fungible Tokens will be continuously recovered! A better strategy is for the market to bundle the payment of ownership transfer fees when purchasing Non-fungible Tokens, so that ownership is transferred to the new buyer at the time of sale. Using this model, supporting royalty payments is closely related to ensuring a better market experience.

Neither the right of recovery mechanism nor the allowlist and blocklist mechanism can prevent the circumvention of royalties by packaging Non-fungible Tokens—unless you prohibit all Smart Contract holdings of Non-fungible Tokens, which is very restrictive (especially considering the rise of account abstraction).

Under the recycling right mechanism, the packaging contract must pay the ownership transfer fee to obtain ownership, thus becoming a legitimate packaging Non-fungible Token. This actually becomes the exit fee, which is the price of leaving the Non-fungible Token ecosystem. In addition, if a popular packaging contract appears, it is easy to identify the contract on-chain.

Any owner of ownership is considered to be a Non-fungible Token of a maliciously packaged contract, which can be prevented from participating in the ecosystem, community activities, or other related utilities of Non-fungible Token by the Non-fungible Token creator. If a packaged contract is identified and blocked from the community, and a Non-fungible Token owner wants to “re-enter” the ecosystem, they can pay a fee to transfer ownership from the packaged contract as a re-entry fee.

More broadly, it may be beneficial to reveal whether the asset owner is also the owner of the ownership. Reducing access to non-owners throughout the ecosystem may become an important incentive for Non-fungible Token buyers to pay royalties. For example, if a market or Wallet prominently displays Non-fungible Token that has not paid royalties/ownership transfer fees, it may encourage consumers to choose to pay royalties.

Assumption

The recycling right framework relies on two key assumptions:

1. The transfer fee for ownership accepted by the creator becomes “royalties”, and royalties are no longer a direct percentage of the sale price.

2. Creators accept that their Non-fungible Tokens may be wrapped to potentially circumvent royalties (knowing there are still costs for exiting and re-entering), and can easily identify and block community access to wrapped Non-fungible Tokens.

If the creator does not accept these assumptions, the right of recovery design cannot exist independently. There are also other functions and components that can relax these assumptions, and we hope to expand on this in the future - we also hope that others in the community can expand on these issues, as we collectively work to address this important problem.

We also recognize that the concept of ownership has deviated from the existing mindset of Non-fungible Tokens. However, similar ownership structures already exist today for Non-fungible Tokens (such as the ENS with registrants and controllers).

Conclusion

When designing the NFT royalty solution, we believe the entire industry is working towards the same goal: maintaining composability, preserving digital property rights, and ensuring that creators are fairly compensated for creating wonderful works.

As more long Non-fungible Token use cases emerge - from collectibles to digital-physical hybrids (digi-fizzy) - there is no one-size-fits-all solution. Every creator (and every Non-fungible Token) is different. Builders and creators should have a simple way to understand various royalty designs and their trade-offs to choose the one that best fits their unique goals. The more we expand the design space, the better the effect.

This industry has the ability to significantly improve the way creators make a living from their work, and perhaps the best way is yet to come. Royalty execution models are new and many are still experimenting with them. If you have any novel ideas after reading this article, please share them with us!