An open letter to SEC Commissioner Peirce on token safe harbors

Gabriel Shapiro is a Silicon-Valley-based attorney specializing in complex corporate and blockchain-based transactions. 

Dear Commissioner Peirce,

I am a California attorney who has been active in representing technology clients in complex transactions for over ten years. After eight years primarily devoted to mergers & acquisitions transactions at the Silicon Valley branches of Dewey & LeBoeuf, Weil Gotshal and Hogan Lovells, I began focusing my work on clients building blockchain technology.

Since then, I have also been very active in writing on the intersection of law and decentralized systems[1], and, among other things, played a significant role in crafting Wyoming’s 2019 corporations code amendments[2] and a security token protocol for Ethereum[3].

I read with interest your recent “Token Safe Harbor Proposal” dated February 6, 2020, and have a number of thoughts thereon that I am eager to share.

In the open-source spirit of blockchain technologies, I will be posting this letter publicly at or about the same time as I send it. However, should you be so gracious as to respond, I will hold your response and all follow-up correspondence in confidence unless you consent to publication. 

The Commission Understands Blockchain And Is Poised To Provide Clarity

The Staff’s doctrine of “sufficient decentralization,” first articulated in William Hinman’s speech “Digital Asset Transactions: When Howey Met Gary (Plastic)” represents a watershed moment in the history of U.S. securities laws, and evidences the Commission’s deep study and understanding of decentralized blockchain technologies. When a blockchain network is “sufficiently decentralized,” it becomes, in effect, a public commons, and the protections of securities laws are thus neither necessary nor desirable. 

Likewise, I applaud and embrace your leadership in driving regulators and legislators to provide developers and investors with greater ex ante regulatory certainty to guide their efforts. The Commission can and should clarify the circumstances in which a blockchain network is sufficiently decentralized. Your proposed safe harbor is an important step toward this goal. 

Because of the important work you, Director Hinman and others have done to lead the way, the Commission is poised to craft a policy that balances the concerns of protecting ordinary markets against an appreciation for the unique promises and nature of blockchain technology. 

Iterating On The Safe Harbor

With this letter I aim to further the goals of your safe harbor while also suggesting some variations that I believe would align it further with traditional U.S. securities doctrine.

U.S. securities laws have evolved in accordance with a time-honored tradition of elevating substance over form. In that spirit, the safe harbor is premised on a technology-neutral determination of when a security is associated with or represented by a token. The safe harbor does not deny that tokens may represent securities, but provides blockchain developers with a specific roadmap for securities law compliance without incentivizing market participants to prefer token-based financings to traditional ones. 

A revised version of the safe harbor is available here. I believe this proposal should be considered. The remainder of this letter explains the most important features of the revised safe harbor and explains the reasoning behind each feature. 

Under this approach, the initial development team may utilize traditional exemptions such as Rule 506(c) or a qualified Tier II Regulation A+ offering to initially sell any pre-mined tokens. While the network is on the path to sufficient decentralization, the tokens may trade as securities on a peer-to-peer basis through decentralized exchanges, on appropriately registered centralized cryptosecurities exchanges or on foreign exchanges which exclude U.S. persons and are based in jurisdictions which do not treat the tokens as securities. Section 12(g) reporting is tolled by modifying Regulation A+ so that it no longer requires the use of a transfer agent for equity securities. 

Before “Sufficient Decentralization", Open Network Tokens Typically Represent Securities 

Before we delve into the details of the safe harbor, it is first important to establish a shared frame of reference for our securities laws as applied to tokens.

For the sake of brevity, I have refrained from a lengthy and citation-heavy discussion in the style of a law review article or memorandum of points and authorities. Of course, I would be happy to provide more detail at a later date. 

1. Nature of Open Network Tokens

In a recent enforcement action, the SEC described how open network tokens may function as “the representation of the holder’s right to the value underlying it, the mechanism for the [token purchasers] to realize their profit.”[4] I endorse this perspective. 

Purchasers do not primarily think of open network tokens as being similar to real estate, consumer products, software licenses or perishable goods such as whiskey or oranges. Nor are such tokens the “object” of an investment scheme, only momentarily twisted into serving an investment purpose against their normal nature.[5] 

Rather, open network tokens—finite in supply and specific to a deployed instance of a software protocol—are mechanisms for sharing in the value of an open network and are designed to appreciate indefinitely as other participants work to make the network more valuable.[6] Such tokens represent a major financial innovation—a method of investing in an ownerless digital commons—and are best understood when considered on their own terms rather than reasoning by analogy. 

The purchase of open network tokens in material amounts will nearly always be primarily or exclusively for investment. We need only ask whether the network is controlled by a relatively small group of affiliated persons (in which case such investment is a securities transaction) or whether the network is decentralized (in which case the securities laws do not apply). 

2. Open Network Tokens As Investment Contracts

When such a token is purchased, whether directly from the development team or in secondary transactions, by a person who has a reasonable expectation of profiting predominantly from the efforts of the initial development team, and there is (by virtue of the initial development team’s continuing economic control or functional control of the network) a common enterprise, our securities laws entail that an investment contract between the development team and the purchaser exists as a matter of law. The contents of the investment contract are the promises made by the initial development team to undertake efforts that implicate the Howey test. The promises may be implied, as with a whitepaper setting forth a vision of the thriving token economy, or they may be explicit, as embodied in a written contract in which the initial development team agrees to build and promote such a network. 

In an extreme case, a development team may have been very careful to refrain from making any promises, express or implied. The development team may even have expressly disclaimed all promises and undertakings, and may have secured from initial token purchasers a representation that such purchasers have not received and are not relying on any promises from the initial development team.[7]

Nevertheless, the development team may still reasonably be expected to be the predominant driver of token profits based on circumstances relating to its operational or economic control of the network. Because our securities laws are non-waivable[8] and elevate substance over form with reference to the canons of construction applicable to remedial legislation[9], even in such “no-promise” circumstances, an investment contract may be deemed to exist between the token purchasers and the development team as a matter of law. 

3. Open Network Tokens As Stock or Other Securities

Our case law provides that the existence of more conventional securities may, where appropriate, be tested “in terms of their substance (the economic realities of the transaction), rather than their form.”[10] Thus, open network tokens may also constitute other types of securities beyond investment contracts.

For example, tokens on “proof-of-stake” networks share many characteristics of “stock”, including the right to receive dividend-like staking rewards, negotiability, pro rata voting rights and the capacity to appreciate in value.[11] Conceptually speaking, such open network tokens function very much like shares of “network equity”[12] or shares in a “decentralized autonomous corporation”.[13]

4. Some Open Network Tokens May Be Non-Securities at Genesis

Some open network tokens may not be securities even when the network first launches, because the Howey test is not satisfied and the policy concerns underlying the securities laws are not implicated. This will typically (but not always) occur when there has been no “ICO,” “IEO” or other sale of “pre-mined” tokens. This is not, however, because persons only receive tokens when the network is functional and a scintilla of functionality is sufficient to moot the securities laws. Rather, it is because such networks are “sufficiently decentralized” at genesis. 

For example, “fair-launched” networks like Grin essentially resemble a kind of general partnership in which token ownership confers a quasi-equity interest in the network, but the securities laws are not implicated because all participants have a highly informed, active managerial role and are not relying predominantly on the entrepreneurial efforts of any other person(s).[14]

Launches bearing similarities to that of ZCash, in which the network is developed, as it were, “on-spec” and the initial developers are paid a reward of tokens from each block, like miners, may also fall outside the ambit of the securities laws, assuming that the developers are very clear that such tokens are ‘payment for work done’ and act consistently with that position (i.e., are not continuing to promote and improve the network in a manner that would create Howey expectations). 

5. Purpose of the Safe Harbor

Regardless of whether open network tokens represent investment contracts, “stock” or other securities, as long as they do not coincidentally also represent unrelated equity or debt rights, the securities laws do not require that they be regulated as securities beyond the point of “sufficient decentralization” of the network.

At that point, there is no longer an issuer or other controlling person that could rationally satisfy the reporting requirements of our disclosure-based securities laws; moreover, most of the risks contemplated by the securities laws are no longer relevant. Although a profit motive (rather than a “consumptive” motive) still prevails, at that point the network is being operated rather like a general partnership in which each participant is relying only on its own efforts, or in any event is not relying predominantly on any other participant(s)’s efforts, to bootstrap the network and achieve profits. 

The primary purpose of the safe harbor is to thus define when this point—the point of “sufficient decentralization”—is reached. As such, it does not revise or suspend the securities laws, but rather turns “sufficient decentralization” from a gloss on the fourth prong of Howey into an objective test providing reasonable ex ante clarity for when the securities laws will cease to apply to an open network token. 

Overview of Revised Safe Harbor

1. Disclosure Statement

The initial development team’s disclosures regarding the project must be filed with the Commission and include specific contractual undertakings (covenants) for developing and promoting the blockchain technology and network. In effect, the Disclosure Statement must set forth the investment contract between the initial development team and the token buyers who are investing in reliance on the initial development team’s efforts. 

Requiring the initial development team to commit to specific contractual covenants has the benefit of providing token holders with concrete rights and enforceable expectations, while also enabling the initial development team to eventually demonstrate that it has honored its responsibilities. The initial development team may still preserve flexibility to experiment, learn and iterate in accordance with the principles of agile software development by building reasonable discretionary standards into the covenants and enabling token holders to approve material changes to the covenants with a token-based or other reasonable governance mechanism. 

2. Network Maturity

Network Maturity is achieved when the initial development team: 

• has performed in all material respects the investment contract associated with the tokens;
• does not economically control the network (owns less than 10% of tokens); and
• does not operationally control the network (owns less than 10% of the means of determining network consensus).

3. Network Maturity as Performance of the Investment Contract

The initial development team’s fulfillment in all material respects of its promises to token buyers is one important aspect of measuring Network Maturity.

To the extent that promises go unperformed, there may remain an investment contract associated with the tokens requiring them to be treated as securities.

Performance of the investment contract will tend to coincide with achievement of “sufficient decentralization”—after all, what the initial development team promised to build was a decentralized blockchain network!

4. Network Maturity as Lack of Economic Control

If the initial development team still owns a material percentage of tokens, token buyers may reasonably expect the team’s efforts to remain the predominant factor affecting token price. Armed with substantial network-specific expertise and a token war-chest, the initial development team has both the power and incentive to increase token value, and token buyers will seek to profit from the team’s efforts.

Thus, even in the absence of any promises, the Howey test can still be satisfied if the initial development team owns substantial tokens. 

Short-swing trading, insider trading and other manipulative or self-interested trading by a major token holder also threatens the economic integrity of the network. The initial development team’s power to engage in such economic attacks is a form of “economic control” that enables coercion of other network participants, unfairly swaying the outcome of contentious hard forks and other governance disputes. The securities laws are a bulwark against such manipulative and self-interested behavior by asymmetrically powerful network participants and should continue applying until the network is sufficiently decentralized. 

The threshold of 10% ownership of existing tokens has been selected to track Section 16’s percentage threshold for defining an “insider” of a public company. In theory, a lower threshold (5%) or a higher threshold could be selected (20%-50%). However, a safe harbor should likely err on the side of being conservative. 

5. Network Maturity as Lack of Operational Control

As you know, a blockchain network is operated on a peer-to-peer basis by computers running a software client which processes transactions in accordance with a set of consensus rules. The consensus rules determine which set of chained blocks is agreed by nodes to constitute the canonical blockchain for that network. 

Whether the power to determine consensus is based on hashrate, token stake, leader selection or a form of per capita voting, the power of each participant to contribute to consensus on a given network at a given time is measurable. We refer to this as the “consensus power” for a network. 

Ownership of substantial consensus power can be a form of operational control over a network. If the initial development team beneficially owns 10% or more of the consensus power of the network, the network may not be sufficiently decentralized, and securities laws may still apply. 

Operational control of the network creates incentives, expectations and potentials for abuse by the development team. The control of so much consensus power could enable the development team to strongly influence governance decisions, and potentially even to attack the blockchain network for its own benefit. Token buyers may reasonably expect that an initial development team with so much operational power will remain the predominant driver of future token profits, satisfying Howey and triggering the securities laws. 

Of course, in proof-of-stake networks, operational control and economic control may be the same, assuming the initial development team chooses to stake its tokens; in such cases, the tokens may be “stock” and the securities laws may be even more strongly implicated. 

6. DAPPs, DAOs, etc.

DAPPs, DAOs and other Open Networks which do not have their own exclusive blockchain should also be eligible for the safe harbor. The revised safe harbor accommodates this possibility. 

7. Secondary Market Regulations Apply to Securities Intermediaries, Not P2P Transactions

Before sufficient decentralization is achieved, tokens will be treated as representing securities, and both primary and secondary market transactions in tokens will be regulated under the securities laws. Thus, the Exchange Act may also apply to such tokens. 

Thus, in tandem with adopting the safe harbor, the Commission should clarify how regulations pertaining to secondary markets in securities apply to open network tokens prior to the point of Network Maturity. The Blockstack Regulation A+ offering of STX tokens provides an excellent starting point for such clarification, and my proposal here takes inspiration from many of the positions Blockstack adopted in pursuing qualification of that offering.[15] 

Blockchain is intended to be a technology facilitating peer-to-peer, disintermediated transactions in tokens. Most secondary market securities regulations are aimed at securities intermediaries. This happy dichotomy suggests the ideal path forward for the Commission: 

• clarify and confirm through guidance the legality of pure, off-exchange, disintermediated P2P trading of tokens representing investment contracts; 
• clarify and confirm through guidance that blockchain network nodes will not be considered broker-dealers or exchange operators solely by virtue of receiving tokens as “block rewards” from the network for transaction validation and block production activity;
• clarify and confirm either that:
  • investment contracts will not be considered “equity securities” for purposes of Section 12(g) of the Exchange Act or 
  • a Regulation A+ offering of a token representing an investment contract will not require reliance on a transfer agent in order to toll the application of 12(g)[16] 
• clarify and confirm that the third-party clearing rules for ATSs do not apply to trades settled on-chain (and thus which do not involve counterparty risk posed by the trading parties to one another); and
• clarify and confirm through guidance that smart-contract-based “decentralized exchanges” that are truly autonomous, non-custodial and peer-to-peer, such as those implementing the Uniswap and 0x protocols, are not required to be registered as securities exchanges or ATSs solely by virtue of the fact that persons may utilize them to trade in tokens representing investment contracts.

Under my proposal, centralized, custodial cryptocurrency exchanges facilitating trades in tokens representing investment contracts, as well as persons acting as conventional broker-dealers for such tokens, would remain subject to securities regulations. This is appropriate, since in the current, very lightly regulated environment for such exchanges, fraud, abuse, loss of funds, front-running, wash-trading and other harmful activities are alleged to be commonplace. Subjecting such intermediaries to regulations and reining-in their bad behavior is not harmful to the intended purposes and functioning of blockchain technology. 

Similarly, aspects of the Exchange Act like Section 16 and Section 13 should be applicable to open network tokens prior to Network Maturity, in order to prevent self-serving exercises of economic control by outsiders seeking to engage in hostile takeovers of the network and insiders seeking to profit from information asymmetries. The recent WeWork scandal, wherein self-dealing on the part of certain of the company’s executive officers was only exposed when WeWork needed to disclose its insider transactions in SEC filings as a prelude to gaining access to the public capital markets, is merely the most recent reminder that “sunlight is the best disinfectant” against toxic governance.

Blockchain development teams are no less prone to such self-dealing than corporate ones.

Similarly, open networks prior to reaching Network Maturity may also be susceptible to the kind of “greenmail” and other hostile acquisition attacks which the Williams Act was designed to prevent, particularly when the consensus power of such open networks is based on proof-of-stake; accordingly, the application of Section 13 may be appropriate. 

No Three-Year Exemption Period

Respectfully, there is one aspect of the prior version of the safe harbor which I find particularly objectionable: the three-year period in which tokens would be exempted almost entirely from the securities laws.

I believe this would be counterproductive. By treating token-based securities financings differently from conventional securities financings, such an approach is unfair both to traditional securities issuers and to participants in blockchain networks that no longer or never did implicate the securities laws.

The essential insight behind Director Hinman’s “sufficiently decentralized” gloss on Howey is that decentralized networks are special and function as a public commons not implicating the securities laws. Blockchain tokens that are decentralized and autonomous represent a public good which is extremely difficult to create and the differences of which from ordinary, centrally managed technologies should be respected by our law. 

1. Three-Year Exemption Would Distort the Capital Markets with Regulatory Arbitrage

Companies like Kik and Telegram, both of which the Commission is currently litigating against, have essentially used token offerings as non-equity-dilutive alternatives to traditional capital-raising. In the first case, ordinary retail investors suffered massive losses, and in the second case, the Commission is seeking an injunction to prevent similar losses. There is no reason why the government should regulate such capital-raising any differently from other corporate financing methods.

Doing so would create a perverse incentive for all companies to use token-based financing merely in order to keep pace with competitors who would face a lighter regulatory load because they raised money by selling tokens instead of stock. Blockchains are cool, but they are not so cool that the government should subsidize their creation by means of regulatory arbitrage. 

2. Three-Year Exemption Would Encourage Meritless Copies and Forks Which Devalue Existing Token Economies

Additionally, there are many forks and copies of decentralized blockchains which offer little in the way of innovation, but have disrupted existing networks for the gain of a very few individuals who re-centralize control over the new chain. A three-year safe harbor would only embolden such value-grabbing initiatives, which would be able to operate for a period of three years without the burdens of securities laws—long enough to turn every member of the development team into someone very wealthy, even if the project ultimately fails and many retail investors are harmed. 

In general, any safe harbor should still respect the specialness of existing sufficiently decentralized blockchain networks and the self-sacrifice and financial restraint exercised by development teams utilizing a “fair launch” which never implicated securities laws in the first place. The participants in such blockchain experiments have endured many trials and tribulations in the pursuit of truly decentralized solutions. 

For example, the development team behind Ethereum decentralized the network, and thus transitioned ETH from security to a non-security, by gradually abnegating its own power and spreading funding among various independent development teams. Such decentralization comes at the cost of speed, coordination, marketing and other benefits[17], and ETH holders have borne that cost for years.

As another example, the developers of the fair-launched blockchain Grin worked on a largely uncompensated basis to create and periodically update the network, in part so that the securities laws would not be implicated in Grin. 

3. Three-Year Exemption Is Not Necessary to Achieve Sufficient Decentralization

The three-year period is not necessary to achieve decentralization. First, it is worth observing: securities are actually rather good at trading and have a reputation for doing so!

However, as I suggested in the prior section of this letter, the Commission can do some things—far less dramatic than creating a three-year open season on “HoweyCoins”—to help improve the state of cryptosecurities rails. 

As a non-exhaustive list: The Commission could allow a blockchain to serve the role of a transfer agent in Regulation A+, could clarify that true uncontrolled DEXs which merely facilitate P2P trades are not required to register as securities exchanges/ATSs and could provide appropriate relief to the third-party clearing rules for ATSs that are utilizing instant on-chain settlement. More centralized exchanges also can, should and will build out properly regulated and licensed cryptosecurities rails to further facilitate such trading. 

Thus, open network tokens can trade very freely, as securities, for years, and sufficient liquidity can be achieved without a three-year carveout to the securities laws. 

Conclusion

I thank you sincerely for considering my thoughts and for your continuing advocacy on behalf of blockchain technologies. I would welcome the opportunity to work with you, the other Commissioners and/or other members of the Staff in determining how best to advance this cause with new rules, regulations and laws.

Of course, I am but one voice on these topics.

I would urge the Commission, the Commissioners and other Staff, both in their official and personal capacities, to participate in the open-source culture of blockchain by doing as much of their thinking on these subjects in public as reasonably possible. 

Endnotes

1. A collection of my thought pieces may be found at https://lex-node.github.io/sustain/projects.html.
2. https://www.wyoleg.gov/Legislation/2019/HB0185.
3. https://github.com/zerolawtech/ZAP-Tech/blob/master/docs/papers-and-research/ZAP-Whitepaper.md.
4. SEC Memorandum of Law in Support of Application for Preliminary Injunction, SEC vs. Telegram, https://www.scribd.com/document/444754972/SEC-MEMORANDUM-OF-LAW-1-IN-FURTHER-SUPPORT-OF-ITS-APPLICATION-FOR-A-PRELIMINARY-INJUNCTION-AND-2-IN-OPPOSITION-TO-DEFENDANTS-MOTION-FOR-SUMMARY-J.
5. Cf. Lewis R. Cohen, Ain't Misbehavin': An Examination Of Broadway Tickets And Blockchain Tokens, 65 Wayne L. Rev. 81 (2019).
6. See Gabriel Shapiro, Size Does Matter – Part 1: A Philosophy of Securities Laws for Tokenized Networks (https://medium.com/coinmonks/size-does-matter-part-1-9f83b130a451).
7. See SEC Order – AirFox (https://www.sec.gov/litigation/admin/2018/33-10575.pdf): “The terms of AirFox’s initial coin offering purported to require purchasers to agree that they were buying AirTokens for their utility as a medium of exchange for mobile airtime, and not as an investment or a security[…]Despite the reference to AirTokens as a medium of exchange, at the time of the ICO, investors purchased AirTokens based upon anticipation that the value of the tokens would rise through AirFox’s future managerial and entrepreneurial efforts.”
8. See e.g. Elaine A. Welle, Freedom of Contract and the Securities Laws: Opting Out of Securities Regulation by Private Agreement, 56 Wash. & Lee L. Rev. 519 (1999), https://scholarlycommons.law.wlu.edu/ wlulr/vol56/iss2/5
9. See Tcherepnin v. Knight, 389 U.S. 332, 336 (1967): “[W]e are guided by the familiar canon of statutory construction that remedial legislation should be construed broadly to effectuate its purposes. The Securities Exchange Act quite clearly falls into the category of remedial legislation. [Footnote 10] One of its central purposes is to protect investors through the requirement of full disclosure by issuers of securities, and the definition of security in § 3(a)(10) necessarily determines the classes of investments and investors which will receive the Act's protections.”
10. United Housing Foundation, Inc. v. Forman, 421 U.S. 837 (1975). See also SEC Paragon Coin Order (https://www.sec.gov/litigation/admin/2018/33-10574.pdf): “In analyzing whether something is a security, ‘form should be disregarded for substance,’ Tcherepnin v. Knight, 389 U.S. 332, 336 (1967), ‘and the emphasis should be on economic realities underlying a transaction, and not on the name appended thereto.’ Forman, 421 U.S. at 849.
11. Forman 421 U.S. at 849.
12. See Gabriel Shapiro, Size Does Matter – Part 1: A Philosophy of Securities Laws for Tokenized Networks by Gabriel Shapiro (https://medium.com/coinmonks/size-does-matter-part-1-9f83b130a451).
13. See e.g. Stan Larimer, Bitcoin and the Three Laws of Robotics (https://letstalkbitcoin.com/bitcoin-and-the-three-laws-of-robotics) and Daniel Larimer, Overpaying For Security by (https://letstalkbitcoin.com/is-bitcoin-overpaying-for-false-security).
14. See e.g., Friendly Power Co. LLC, 49 F. Supp.2d at 1369.
15. See https://www.sec.gov/Archives/edgar/data/1719379/000110465919020748/a18-15736_1ex1a15addexhbd6.htm.
16. The Staff should also consider liberalizing Regulation A+ in additional ways (such as by increasing the public float ceiling under the 12(g) exemption), but should do so across the board for all Regulation A+ offerings, not merely tokens.
17. Ben Edgington, Ethereum’s ‘Bazaar’ Development Model Will Pay Off in 2020 (https://www.coindesk.com/ethereums-bazaar-development-model-will-pay-off-in-2020).