Liquid Staking Legal Considerations – Condensed Overview

Disclaimer: This document does not constitute legal, financial or other advice and is not intended to be relied upon or used by any person for any purpose, other than informational and educational purposes. No attorney-client relationship or privilege is intended to be created or implied. No representation or warranty is being made as to the quality or fitness for any purpose of this document.

Like most uses of blockchain technology, liquid staking involves novel questions of law and entails legal uncertainties. In this section, we provide a brief overview of the most important legal considerations related to liquid staking. We use the term “StakeToken” to refer to a token that represents any kind of liquid staking arrangement, although the exact nature of the StakeToken will vary based on the type of liquid staking at issue. We refer to “validators” as those who have a direct power to validate or produce blocks as part of the native consensus-forming activity of the network, to “stakers” as those who natively or non-natively delegate validation-power to validators by locking up tokens and to “staked tokens” or “staking tokens” as the native tokens that must be natively locked up on the network to exercise validation power. Included as Appendix A is an extended analysis of liquid staking legal issues which explains the reasoning behind our conclusions in greater depth.

1.      Legal Engineering.

The use of legal agreements—contracts written in natural language—may be necessary or advisable for certain forms of liquid staking. We refer to this as a form of “legal engineering.” The importance and aims of legal engineering will vary based on the type of liquid staking at issue:

·        Delegation Exchange. Delegation exchange would likely require moderate legal engineering, with the details depending on the exchange method used. One simple (albeit non-cybersecure) method would be for a user to sell their private key (and all associated usage rights) which controls the staked tokens to another user through a bill of sale or similar agreement. A delegation exchange system like B-Harvest allows stake to be controlled by a group of accounts and traded among such accounts. Such groups might be analogized to partnerships or investment clubs, and may call for different legal engineering depending on the purposes intended by a particular group with respect to a particular pool—e.g., if stakes are to be exchanged among the group members, then contractual covenants will be needed to enforce payment obligations and determine the exact moment when title to the staked tokens changes hands.

·        Native Liquid Staking/Delegation Vouchers. Generally native liquid staking/delegation vouchers will require little to no legal engineering. Participants in native PoS network mechanics may be seen as implicitly agreeing to a kind of code deference arrangement whereby they accept the results of the operation of the PoS network, including any applicable native liquid staking mechanics. In many PoS networks, validators and stakers have no direct social or contractual relationship with one another, but only interact indirectly through the protocol; as a result,, they have little or no ability to cause harm to one another and  the risk of disputes between them is low. In such contexts, legal engineering is not required, although validators may still wish to publish some general disclaimers of liabilities. Even when a validator and its stakers do have a relationship, such as in Tezos (where validators must directly distribute staking rewards to stakers), the relationship may be a relatively conventional commercial relationship defined by a fairly simple written terms of service.

·        Non-Native Liquid Staking. Non-native smart contract liquid staking systems may call for different forms of legal engineering, depending on the specifics of the solution and how much flexibility or governance it contemplates. The core dynamics of Stake DAO’s StakeTokens (LTokens) are rather simple and predictable and afford little discretion to anyone—the LTokens function like notes or certificates of deposit that can be turned into the smart contract to receive stake and staking rewards—thus, a kind of ‘code deference’ approach might be possible, and legal engineering might not be needed. On the other hand, the interest-rate features of Stake DAO and the rules for the DAO itself may require legal engineering. Systems like Everett and Acala that attempt to make the StakeToken (“bAtoms” for Everett, “L-DOTs” for Acala) do double duty as a stablecoin pegged 1:1 with the underlying staking token (“Atoms” for Everett, “DOTs” for Acala) will present a more complex risk profile that could implicate legal engineering concerns more strongly. Because such systems will presumably be governed by token holders proactively (to hold the peg), the obligations of the governors and the rights of the StakeToken holders should be carefully defined in a terms of service or similar legal agreement. Systems like StakerDAO, whose StakeTokens will likely be securities governed by representatives of token holders, will obviously require extensive legal engineering to define the rules of representation and collective action, provide indemnification and insurance to the representatives, and so on. In all cases, the complexity and novelty of non-native liquid staking systems means the requisite legal engineering will not be as simple as copy-pasting boilerplate terms of service for a simple consumer-facing website. Legal engineering may also be required to deal with some of the extra regulatory risks involved in non-native liquid staking (see below).

·        Custodial Liquid Staking. Custodial liquid staking is inherently tied to contract law. In order for a custodian (like a traditional cryptocurrency exchange, but also like RocketPool in its initial version) to hold tokens on behalf of a staker, there must be a legal agreement which defines the parties’ respective rights and obligations as part of that relationship. The law provides for different types of custody—trusts, bailments, escrows, etc.—and custodians should take pains to define which type applies to the staking tokens entrusted to the exchange for staking. Similarly, since the StakeTokens are not defined by a decentralized protocol, legal engineering is required to define them. Are the StakeTokens transferable legal instruments, i.e., “a form of electronic title document…represent[ing] a record of title to” the staked tokens and associated awards? [1] Or are they a mere certificate of deposit from the custodian? Or something else? Who bears the risk of loss of tokens due to slashing events—the StakeToken holder or the custodian? Does the answer differ depending on the validator’s level of fault (negligence vs. gross negligence vs. recklessness) or on the type of slashing event (double-signing vs. downtime)? For custodial liquid staking, legal engineering will be required to answer all of these questions and more.

·        Synthetic Liquid Staking. Synthetic liquid staking will feature a ‘swap contract’ defining the terms and conditions of payouts based on market events affecting the value/price of the staked tokens or slashing of the staked tokens. This will initially require significant legal engineering, but over time certain standard forms of such swap may become prevalent and reduce the expense of legal engineering.

2.      Regulations.

The application of regulations to PoS networks is still generally poorly defined and tested. Even regulatory agencies such as the Securities & Exchange Commission (“SEC”), Commodities Futures Trading Commission (“CFTC”) and Financial Crimes Enforcement Network (“FinCEN”) which have been relatively proactive in tackling blockchain regulatory issues generally, have nevertheless failed to provide significant guidance specific to PoS; instead, their guidance typically discusses Bitcoin or Ethereum 1.0, which are proof-of-work (“PoW”) networks, despite the fact that there are many potentially legally important distinctions between the two types of networks. Accordingly, there remains a much greater “gray area” for how regulations apply to PoS networks than to PoW networks.

Liquid staking for PoS involves the creation of derivative instruments on top of PoS, and thus adds even more legal complexity and uncertainty to PoS. Nevertheless, if certain forms of liquid staking are more likely to carry heavier regulatory burdens than others, it is important to understand and wrestle with that risk as early as possible in the game, since it may inform where people should best apply their efforts to develop liquid staking and avoid misallocation of capital into types of liquid staking that are non-starters from a regulatory point of view.

·        Delegation Exchange. Delegation exchange would likely not be subject to significant regulations. In the B-Harvest system, depending on the details of how a particular group account uses a particular pool, some commercial and contracting regulations could be implicated, but they are likely to apply in a very similar way to how they apply to other situations where people get together and co-manage assets. Of course, if the group is very large and trading in staking positions among the group members is extensive, regulations regarding securities exchanges, commodities exchanges, commodities pools and/or investment funds could be implicated; however, considering that these groupings are inherently short-lived because they apply to a specific set of staked tokens, and the underlying staking interest is itself likely to be lightly regulated, it is unlikely such group accounts would ever reach sufficient scale to implicate these regulations.  

·        Native Liquid Staking. Native liquid staking will generally be the least likely to implicate regulations. No one in particular controls or has the power to change the blockchain network or protocol supporting native liquid staking functionality, and thus there is no natural target for regulation. When relying on decentralized blockchain systems, users may face risks from poor software design, but generally do not face the kinds of risks that require heavy regulations—risks from custodians, fiduciaries and similar types of intermediaries who are most apt to abuse information asymmetries or pose other risks based on conflicts of interest. Moreover, the development and deployment of software in and of itself is heavily protected in countries like the U.S. by freedom of speech principles, meaning that governments may lack the power to regulate such systems heavily without violating civil rights. However, there is some residual risk of regulations applying even to native liquid staking; for example, we cannot completely rule out that in some PoS networks the native token could be a security and thus the StakeToken could be a securities derivative.

·        Non-Native Liquid Staking. Actively governed smart contract systems for liquid staking may face higher regulatory burdens:

1.      The stakers may be seen as having “loaned” the staking tokens to or deposited the staking with ‘the system’ or the unincorporated association of governance token holders, in which case the StakeTokens could represent a kind of promissory note or certificate of deposit. This characterization will be more likely for systems like Stake DAO, which tie each StakeToken to a specific stake and pay interest to the staker during the lock-up period. Alternatively, the smart contract system may be seen as having “loaned” the StakeTokens to the stakers. This characterization may be more likely for systems like Everett and Acala which follow a MakerDAO-like collateral/foreclosure design pattern. Either perspective may implicate lending/credit regulations, banking regulations or (if the StakeTokens are analogized to bonds) securities regulations, but in general we regard the case for the applicability of such regulations being weaker or the adverse effects of complying with such regulations if they do apply as being less severe than for some of the other regulations mentioned below.

2.      The stakers may be seen as having disposed of/sold the staking tokens (or a portion of the staking tokens, or the associated staking rewards) with an option to re-acquire them at a later date in exchange for the StakeTokens, which may under certain circumstances be considered a CFTC-regulated option or a CFTC-regulated “swap” (note: many commodities options may also be swaps, so the two categories are not mutually exclusive). From this point of view, the smart contract system embodies an option, with the StakeToken being the “exercise price” required to receive the extra collateral and the staking rewards. Swaps are required to trade only on CFTC-registered “designated contract markets”. However, non-native liquid staking systems do have some noteworthy differences from options, particularly if they also involve the payment of a variable stability fee (which is less like an option exercise price and somewhat more similar to payments of interest on a debt) and swaps (particularly because title to the staking tokens and potential rewards moves with the StakeTokens, rather than merely risk). Even if non-native liquid staking systems do facially meet the definition of “swaps,” it is at least possible that one of the exceptions may apply in at least a subset of cases, such as the exception for “commercial merchandizing transactions” involving deferred delivery’. These details would require extensive analysis that is beyond the scope of this report.

3.       

• The stakers in overcollateralized stablecoin systems like Everett and Acala may be seen as seeking to acquire the future staking awards (and extra collateral) on a “leveraged, financed or margined” basis. From this perspective, non-native liquid staking may be considered a type of CFTC-regulated retail commodities transaction that is required to occur only on CFTC-registered “designated contract markets” unless all participants are “eligible contract participants” (i.e., generally having $5M-$10M in assets).
• The terms “leverage”, “finance” and “margin” have not been strictly defined for purposes of the applicable regulations, and the CFTC has interpreted them broadly in the past so that “indebtedness in the traditional sense (i.e., the use of borrowed money) is not required.”[2] Indicia of leverage are said to include “allowing a customer to control a large amount of a commodity with a comparatively small amount of [another commodity]” and “allow[ing] customers to significantly boost their profits with a relatively small investment while also magnifying their losses”.
• Although the StakeTokens are “overcollateralized” in a certain sense (since users will presumably deposit more of the staking tokens than they receive in StakeTokens), when the potential staking rewards and slashing events are taken into account, the can be seen as “leveraged”: after all, it could become “insolvent” in certain situations (massive slashing events, or a “black swan” event like the auction efficiency failures observed in MakerDAO’s “Black Thursday,” etc.). Expressed another way, the locked-up collateral is both the “margin” for obtaining the potential staking rewards and the “margin” for the associated stablecoins that are supposed to have a 1:1 peg—when all of that value is taken into account, the system may be seen as having the same of kind of capital-multiplying effects as more conventional margin trading, notwithstanding the apparent “overcollateralization”. The governance token holders may also be seen as providing “financing” to the stakers, since they are (or are at least often presented as) “lenders of last resort” charged with holding the peg by diluting themselves in insolvency events.[3]
• Because the arguments for regulations applying are fairly persuasive, and the consequences of such regulations would be devastating, engagement with the CFTC should be a priority for teams involved in these systems. Their goal will need to be convincing regulators either that: (a) the decentralized, automated/autonomous or transparent nature of the system eliminates the risks that make regulation appropriate when they occur on a centralized basis; or (b) the regulations should be modified to allow such systems to be registered with the CFTC as regulated designated contract markets.  

4.      Depending on the extent of their governance powers, the association of governance token holders involved in certain of such systems may be seen as performing a money services business role by accepting staking tokens from stakers and depositing them with / for the benefit of validators, or may be seen as subject to other custodial regulations. If so, there may be a requirement that the system be registered with FinCEN or other regulators, perform KYC checks on depositors, etc.—which obviously would be very challenging for a smart contract system.

5.      The governance tokens in systems like Everett and Acala, or the StakeTokens themselves in systems like StakerDAO, may be seen as securities, particularly if a particular entity or group of persons holds a large percentage of the governance tokens and has principal responsibility for promoting or developing or maintaining the system; in such a case, the governance tokens may be limited in the venues on which they can trade and those who hold them may have securities disclosures obligations. This would be a particularly challenging result for governance tokens when they are minted as a source of last-resort funding when the system is insolvent—such issuances would be new securities sales which must be either registered with the SEC or exempt from registration (the latter being unlikely in a decentralized transaction).

6.      The StakeTokens may be seen as bearer instruments subject to certain tax penalties, especially when they are analogous to bonds or other kinds of debt instruments.

·        Custodial Liquid Staking. Custodial liquid staking will in general present a risk of the custodian being subject to money services business regulations—similar to any cryptocurrency exchange business. However, such regulations may be less likely to apply if the exchange itself is also a direct validator and thus is not transmitting funds on behalf of the stakers to a third party. Custodial liquid staking could also implicate the same commodities regulations as were mentioned immediately above for non-native liquid staking; if so, then the exchange would likely need to become a registered CFTC-regulated exchange. Custodial liquid staking will also present other similar regulatory concerns as mentioned above for non-native liquid staking—but the case for them applying will be even stronger, because the exchange is more clearly an ordinary centralized business of the kind usually targeted by such regulations. On the other hand, a traditional centralized custodial like an exchange can also respond to such regulatory issues more nimbly. For example, an exchange could deliver staking tokens back to U.S.-based stakers every 28 days and thus avoid being subject to the CFTC’s regulations of retail commodities transactions involving leverage, margin or financing. Likewise, an exchange can easily impose KYC requirements and thus lock out U.S. investors from liquid staking entirely.

·        Synthetic Liquid Staking. Synthetic liquid staking is simply a commodities swap and thus would be subject to all of the swap regulations under the Commodities Exchange Act as amended by the Dodd-Frank Act, including generally being limited to trading only on designated contract markets registered with the CFTC. Again, there may be an opportunity to persuade regulators that the normal policy concerns with swaps should not apply, and therefore that the regulations should not apply, to the extent that the swap logic can be embedded in autonomous smart contracts which eliminate the kinds of counterparty risks that made swaps an existential market threat in the 2007-2008 financial crisis.

[1] “Digital and Digitized Assets: Federal and State Jurisdictional Issues,” American Bar Association Derivatives and Futures Law Committee Innovative Digital Products and Processes Subcommittee Jurisdiction Working Group.

[2] See Andrew P. Cross, Swaps and Retail Commodity Transactions (Leverage, Margin or Financing: Will We Know It When We See It or Only After It Has Been Identified As Such?).

[3] See e.g. The DAI Stablecoin System: Whitepaper: “MKR….serves as a backstop in the case of insolvent CDPs”. See also Reddit comment believed to be from Rune Christensen: “It is unfortunately not possible to change the fundamental feature that MKR holders are on the hook for all the dai in the system - this is because it[’]s a crucial requirement that all dai has to be fungible, so ultimately they have to have the same backing of last resort.”