Thank you to the Blockchain Association Singapore for inviting me to speak at this Blockchain Week event and for the chance to be back in Singapore, albeit virtually, for another conversation about crypto. I miss Singapore’s people, beauty, food, and energy, but the issues we are talking about fit quite naturally into the virtual experience.
I must start, of course, with my standard disclaimer. The views I represent are my own and not necessarily those of the Securities and Exchange Commission or my fellow Commissioners. That disclaimer is fitting for today’s remarks, as I will be discussing both crypto and enforcement, subjects on which we are not always perfectly aligned with one another.
During the COVID-19 quarantine in the United States, many people, bored with sitting inside, have taken up new outdoor activities. Walking, running, biking, and even inline skating, which is commonly called Rollerblading, surged in popularity. Inline skating, which was quite popular at one point but had fallen out of favor pre-COVID, is an innovation on the much older sport of roller skating. Instead of four wheels in a rectangular formation, some enterprising person placed wheels in a vertical line to mimic the blade of an ice skate.
By most accounts, this phenomenon of shoes with wheels started with John Joseph Merlin, the eighteenth century inventor of roller skates. Merlin, a serial inventor and maker of machines of all sorts, moved from his native Belgium to Paris and then to London. He decided to wear his newly invented roller skates to a party. For effect, he was playing the violin as he rolled in. Unfortunately, the braking system on his roller skates was undeveloped, which had painful results for him and shattering results for the fancy mirror that stopped him.
Merlin’s story brings to mind the excitement and messiness of innovation. Innovators—in addition to being brilliant—are sometimes enthusiastic showmen who do not spend a lot of time thinking about the things they might break along the way. Had he consulted an objective third party, Mr. Merlin might have been told to put down the violin, skate outside of the vicinity of large mirrors, and build in some braking technology.
A regulator can be that objective third party. Ideally, regulators both allow innovators their skating room and incentivize them to identify and mitigate any harmful consequences. But, too often, regulatory objectivity gives way to other considerations that squash innovation. The regulator deems the innovation too dangerous or insufficiently beneficial and sends the innovator packing.
As a regulator, the SEC is charged with protecting investors, facilitating capital formation, and fostering fair, orderly, and efficient markets. I believe that investor protection includes ensuring that investors have a wide range of investment opportunities so they can build portfolios appropriate to their objectives. I also remind myself frequently that it is not my place to make investment decisions for others. Though it is an obvious thing to say, individuals know the most about their own situation, and therefore are best able to assess their own circumstances and make decisions for themselves and their families. I view capital formation as an essential way to transform individual lives and communities, so I look for ways to make it easier for people with great ideas to find funding. Finally, I place great emphasis on the role that disruptive technology plays in keeping our markets fair, orderly, and efficient.
We Commissioners and the staff who work at the SEC bring a variety of perspectives to our work. The interaction of those diverse viewpoints across the SEC is one of the features I enjoy most about being a Commissioner. As you will hear today, however, I do not always convince my colleagues to sign on to my view. In other words, I sometimes get outvoted. I want to take a few minutes today to explain my disagreement with our enforcement action against Telegram. I share these thoughts in the spirit of transparency and with the hope that they will further my efforts to work collaboratively with my colleagues to provide certainty on how a token distribution can be done without running afoul of the securities laws.
Before turning to Telegram, it is worth reminding ourselves that we bring enforcement actions to hold individuals and companies to account for their violations of the securities laws. Often, as part of punishing people for wrongdoing, we are able to get money back to the people who have been wronged. Many of our enforcement actions, including many “crypto” cases we bring, are aimed at holding people accountable for material misstatements they allegedly made while raising money from investors.
Enforcement actions can be instructive to people other than the wrongdoer, but are not an appropriate mechanism to create new law. Our regulatory integrity demands that enforcement actions be premised on a violation of a clearly articulated statute or rule. Where there is ambiguity, we try to provide guidance before bringing enforcement actions.
Even where there is a violation, we may opt to issue a so-called 21(a) report, rather than bringing a full-blown enforcement action. Section 21(a) of the Exchange Act authorizes the Commission to investigate violations of the federal securities laws and, in its discretion, to “publish information concerning any such violations.” The intent behind these reports is not to punish actors for violations, but rather to shine a light on a particular practice that violates the securities laws. These reports serve as an instructive warning for all market participants about how the SEC applies the law to a particular type of conduct.
The SEC issued such a report after investigating the DAO, a decentralized autonomous organization that was designed to enable individuals to work together collaboratively outside of the traditional corporate form. To achieve this goal, the DAO sold tokens that would permit participants to vote on projects and potentially receive earnings from the projects. The DAO “would create and hold a corpus of assets through the sale of DAO Tokens to investors, which assets would then be used to fund ‘projects’” and the tokenholders “stood to share in the anticipated earnings from these projects as a return on their investment in DAO Tokens.” The DAO Report, applying the framework drawn from the now famed Howey Supreme Court case, found the tokens were securities. Briefly summarized, Howey teaches that, regardless of form, something is a security if it represents an investment in a common enterprise with the expectation of profit solely from the efforts of others. I was not at the SEC when the DAO report was issued and am not sure I would have supported it had I been there. I wonder, for example, whether the “solely from the efforts of others” prong was met in that case. The report significantly downplays the token holders’ ability to control the enterprise. It also considers the pseudonymity and dispersion of the token holders, two built-in advantages of blockchain technology, as evidence that the curators’ efforts were essential to the enterprise.
In any event, the DAO Report offered an important warning to would-be token issuers. It put people on notice that if you raise money for an enterprise by selling investment contracts, the securities offering framework applies. The message was clear:
[I]ssuers of distributed ledger or blockchain technology-based securities must register offers and sales of such securities unless a valid exemption applies. Those participating in unregistered offerings also may be liable for violations of the securities laws. Additionally, securities exchanges providing for trading in these securities must register unless they are exempt.
An appropriate and unassailable warning. The question, however, remained—and in my view still remains: What makes something a “distributed ledger or blockchain technology-based” security?
Almost two years later, in 2019, the Commission staff attempted to answer that question. This time it came in the form of “a framework for analyzing whether a digital asset has the characteristics of one particular type of security – an ‘investment contract.’” This staff guidance identifies thirty-eight separate “characteristics” to consider when analyzing whether an offering of digital assets is likely a securities offering. Although I appreciated the attempt, the complexity of this guidance and resulting public confusion motivated me to suggest a safe harbor.  The safe harbor would allow legitimate token offerings to move forward without having to answer the question of whether the token is a security for three years. During that time, the developers of a token could build the kind of functioning, decentralized network that would push the token clearly outside the securities law framework.
In the meantime, most of the hints at clarification coming out of the SEC have been embedded in enforcement cases. In other words, rather than provide useful guidance on safety standards and functional braking technology, we simply sue skaters for breaking mirrors. The resulting settlement orders and litigation releases leave the industry to guess at the path to compliance. Candidly, given the approach we have taken, I often struggle to see a path to compliance; my colleagues sometimes see the sale of a security when all I see is a sale of tokens to be used in a network. How can a token network ever get off the ground if every token distribution event is viewed as a securities offering?
That brings us to the Telegram case. Last month’s settlement was the unsatisfying culmination of an enforcement action that I did not support from the beginning. Telegram had built an operational network, made good faith efforts to comply with the federal securities laws in raising funds to build that network, and engaged extensively with the SEC staff. With the assistance of sophisticated counsel, Telegram used the Simple Agreement for Future Tokens (SAFT) offering structure, which divides the creation of the network and the delivery of the tokens into at least two distinct stages. First, Telegram raised funds to develop the blockchain technology underlying the Grams by selling interests in the anticipated Grams to accredited investors. In return for providing the necessary funds, the accredited investors would receive an allotment of Grams upon the launch of the TON Blockchain. According to Telegram, this first stage—raising funds from accredited investors in a private offering—involved a securities transaction conducted in reliance on Rule 506(c)—a frequently used exemption from our registration requirements.
It was at the second stage that the sharpest disagreements erupted. In one view, the second stage was to begin with the launch of the TON Blockchain and end when Telegram delivered the Grams to the accredited investors. The accredited investors then could resell the Grams, subject to certain previously agreed-upon lockup restrictions. In Telegram’s view, these resale transactions by the accredited investors would not involve a security, but rather a digital currency, which could be used for buying and selling goods and services on the functional TON Blockchain. As network effects took hold, the value of that digital currency, of course, would go up.
This offering structure is rooted in the idea that a token—even one that is initially sold as part of an investment contract—is not itself a security, but instead an asset that has use and value as a medium of exchange on a functional blockchain network. For this reason, such tokens, once they have a consumptive use, should be able to be sold to purchasers outside of a securities transaction. This approach treats as separate the capital raising necessary to build the platform, which is conducted pursuant to the securities laws, and the resale of a functional token to consumers in secondary transactions after the platform is functional, which is not conducted pursuant to the securities laws.
The district court in the Telegram case, at the urging of the SEC, rejected this analytical approach. In the court’s view, the accredited investors’ resale transactions were an offer or sale of a security because the Gram was an integral part of an investment contract. In reaching this conclusion, the court determined that the entire sequence—the Gram purchase agreements entered by the accredited investors, Telegram’s delivery of the Grams upon launch of the TON Blockchain, and the accredited investors’ resale of the Grams—were a single scheme that constituted an investment contract under the securities laws.
Gone is the distinction between the investment contract (the agreement between Telegram and the accredited investors) and the token (the asset to be created and delivered under the agreement). True, the court states that “the security in this case is not simply the Gram, which is little more than alphanumeric cryptographic sequence,” but it is a challenge to reconcile that statement with the district court’s conclusion that Telegram engaged in a “scheme . . . to offer Grams . . . with the intent and purpose that these Grams be distributed in a secondary public market, which is the offering of securities under Howey.” The premise of this conclusion, and nearly all the preceding reasoning, is that the Grams themselves are securities. Indeed, the SEC said as much in its complaint. Furthermore, in reaching its conclusion, the court did not grapple with the implications of the fact that the alphanumeric cryptographic sequence—the Gram—would be usable on a functioning blockchain network at the time of the distribution. Nor did the court analyze whether the efforts of Telegram post-launch were essential to the usability—and hence the potential for increasing value—of the Grams, such that the Howey test would continue to be met. By focusing the Howey analysis exclusively on the original transaction, the factual reality of the transaction that the court halted was lost.
To support its decision to aggregate the entire sequence into a single scheme, the court looked to the Howey Court’s refusal to treat the land and service contracts at issue in that case as separate transactions. This is true enough; however, it was the promoter in Howey that linked the two contracts together as part of a single scheme to sell interests in orange groves. Nothing in Howey indicates that a subsequent, separate resale of the land containing the row of orange trees—but not accompanied by a service contract—also would be a securities transaction. After all, linking the two distinct contracts together as a means to obtain a profit underpinned the investment contract analysis; severing the link and selling only the land meant that one was just selling a row of orange trees. For this reason, Howey would seem to shed little light on whether the sale of tokens that are usable on a functioning digital network are securities simply because the seller acquired the tokens in a securities transaction.
Finally, one last aspect of the Telegram case bears mentioning. Telegram was not a United States company, and its major operations were not in the United States. Only 39 of the 175 accredited investors were in the United States, and those 39 provided only about one-quarter of the funds raised in the private offering. Notwithstanding these facts, when the SEC sought a preliminary injunction, it asked the district court to enjoin Telegram from delivering Grams to “any persons.” As Telegram discovered when it asked the district court to clarify the scope of the injunction, this meant that Telegram could not deliver any Grams to any purchaser anywhere in the world. Prevented from getting Grams into the hands of the people who wanted to use them, Telegram abandoned its platform. This willingness of the SEC to ask for, and of the district court to grant, such sweeping injunctive relief against a non-US company, in a case where one-quarter of the funds came from US investors, reasonably might raise some concerns among our international colleagues. While the United States certainly has the most robust capital markets in the world, and while we work hard to preserve those markets with a fair and efficient regulatory structure, we would do well to recall that our way is not the only way. We should be cautious about asking for remedies that effectively impose our rules beyond our borders.
After the district court’s decision, and its clarification that its decision applied worldwide, Telegram chose to end its legal battle by settling with us. I did not support the settlement because I did not support the underlying action. I do not support the message that distributing tokens inherently involves a securities transaction. What the SEC’s Telegram complaint cast as evidence of an illegal securities offering—that “the project would require ‘numerosity’: a widespread distribution and use of Grams across the globe,” I see as a necessary prerequisite for any successful blockchain network. Moreover, the settlement produced an unsatisfying result—the abandonment of the project. I also found it ironic that the settlement included $1.2 billion in disgorgement, which will go to repay the initial purchasers, the people the district court determined were an integral part of the securities law violation.
Who did we protect by bringing this action? The initial purchasers, who were accredited investors? The members of the public, many of whom are outside the United States, who would have bought the Grams and used them to buy and sell goods and services on the TON Blockchain? Did they really look to U.S. securities laws for protection? Would-be innovators, who will now take additional steps to avoid the United States?
Innovation in the digital asset industry has and will continue to challenge us as securities regulators. I would prefer that we not only hold accountable the reckless innovators who skate among mirrors while playing the violin, but also attempt to provide the more cautious innovators some guidance on how to avoid the hall of mirrors and on what we consider to be adequate braking technology. I look forward to working with my colleagues to ensure that we do so in a consistent and transparent manner.
 See, e.g., SEC Emergency Action Stops Digital Asset Scam (Mar. 20, 2001), https://www.sec.gov/litigation/litreleases/2020/lr24775.htm (alleging ongoing securities fraud, including allegations that defendants claimed the coin was backed by a $1 billion art collection or $2 billion of gold, and promises of a return up to 224,923%); SEC Charges Convicted Criminal Who Conducted Fraudulent ICO Using a Fake Identity (Jan. 17, 2020), https://www.sec.gov/news/press-release/2020-12; SEC Obtains Emergency Order Halting Alleged Diamond-Related ICO Scheme Targeting Hundreds of Investors (May 21, 2019), https://www.sec.gov/litigation/litreleases/2019/lr24473.htm (alleging a $30 million Ponzi scheme involving false promises of investments in diamonds and cryptocurrencies).
 15 U.S.C. 78u(a)(1).
 SEC v. W. J. Howey Co., 328 U.S. 293 (1946).
 See SEC v. Telegram Group Inc., No. 1:19-cv-09439-PKC (S.D.N.Y. Mar. 24, 2020) (opinion and order granting preliminary injunction) [hereinafter Telegram]; see also Complaint, SEC v. Telegram, 19-cv-09439-PKC (S.D.N.Y. Oct. 11, 2019) [hereinafter Complaint].
 See, e.g., Complaint ¶ 7 (“Unless enjoined, Telegram’s completion of the Offering will allow it to have circumvented the Securities Act’s registration requirements, leaving U.S. investors to buy and sell Grams without the vital information about those securities and about Telegram that Congress intended registration to provide.”).
 See Telegram at 18.
 See Howey, 328 U.S. at 295 (noting that “[e]ach prospective investor is offered both a land sales contract and a service contract, after having been told that it is not feasible to invest in a grove unless service arrangements are made.”).
 Indeed, this reading of the limitations of Howey’s reasoning is consistent with earlier Supreme Court precedent. SEC v. C.M. Joiner Leasing Corp., 320 U.S. 344, 348 (1943) effectively acknowledges that the sale of a property interests—there assignments of oil and gas leases—by itself was not a security. The property interests became investment contracts only because of the promoter’s accompanying commitment, made to the purchasers, to develop the leases.
 See Complaint ¶ 2.
 See Complaint ¶ 12.
 See supra note 12.
 See Complaint ¶ 104.
 The Telegram court found that the accredited investors were statutory underwriters because they “bought Grams from Telegram . . . with an intent to resell them for profit in the secondary market soon after launch of the TON Blockchain.” Telegram at 20. However, the court does not conduct the typical subjective inquiry one would expect when determining whether an individual purchased with a view toward distribution of a security under Section 2(a)(11). See generally Louis Loss, Joel Seligman & Troy A. Paredes, Securities Regulation, Ch. 3.A.3 at n.620. The holding may raise larger questions outside of the SAFT context about the ability of certain purchasers of securities in private placements to rely on the Rule 144 safe harbor for the resale of such securities. See Scott Kupor, Reading Between the Lines: SEC, Telegram, and Rule 144, Andreesen Horowitz (Feb. 22, 2020), https://a16z.com/2020/02/22/reading-between-the-lines-sec-telegram-and-rule-144/.