Thoughts on the SEC’s Initial Coin Offering Bulletin and Investigative Report

Summary: this post focuses on the Investor Bulletin and Report of Investigation issued by the SEC in late July. I examine the meaning of the report, implications for ICOs, and thoughts for moving forward. Disclaimer: this is not investment advice, and I am not a lawyer.

Two weeks ago, the U.S. Securities and Exchange Commission (SEC) provided us with two pieces voicing their take on Initial Coin Offerings.

The first was the Investor Bulletin: Initial Coin Offerings. This was largely an educational article warning investors of precautions and possible downsides of investing in ICOs.

The second was the Report of Investigation Pursuant to Section 21(a) of the Securities Exchange Act of 1934: The DAO (PDF) along with an associated press release.

Everyone was wondering when the SEC would offer an opinion on ICOs. Finally, we have something, but it focused on a specific ICO. We cannot generalize just yet, despite attempts at doing so with sensationalist titles such as “Oh Shit, the SEC Just Ruled That Ethereum ICO Tokens Are Securities” (false, the SEC did not make any such blanket statement).

The bulletin was largely investor education. It explained blockchains, ICOs, and virtual currencies at a high level. It advised investors to watch out for fraud and to be aware of securities law such as accreditation requirements. It talked about limits on recovering funds if something goes wrong due to the nature of blockchain investments. Overall, much of this is not unique to ICOs. The same sorts of things could be said for angel investments in startups.

The Report of Investigation was more interesting. The first paragraph and a few quotes from the report really say it all:

The United States Securities and Exchange Commission’s (“Commission”) Division of Enforcement (“Division”) has investigated whether The DAO, an unincorporated organization; UG (“”), a German corporation;’s co-founders; and intermediaries may have violated the federal securities laws. The Commission has determined not to pursue an enforcement action in this matter based on the conduct and activities known to the Commission at this time.
The DAO was created by and’s co-founders, with the objective of operating as a for-profit entity that 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.” The holders of DAO Tokens stood to share in the anticipated earnings from these projects as a return on their investment in DAO Tokens.
Based on the investigation, and under the facts presented, the Commission has determined that DAO Tokens are securities under the Securities Act of 1933.

From this, we see that the SEC concluded a specific ICO and its tokens operated like a security. The ICO/tokens in question were for The DAO. The SEC also declared they will not pursue enforcement action. Interesting!

As a first order of business, let’s examine the surface level view of the report and their assessment of The DAO. For anyone that is not familiar with The DAO, The DAO was created to pool funds for projects and provide people a way to share in the earnings of those projects. Funds were pooled via an Ethereum contract and tokens were issued to contributors to track their ownership in earnings. As an aside, it turns out that The DAO’s contract code had a bug, and someone exploited that bug to steal funds. While that stinks and was a warning to investors, it was not the point of the SEC report.

This was the point of the SEC report: we know The DAO pooled funds and intended to use those funds for projects, providing people with a stake of the earnings of those projects. Sounds a lot like stock in a company, or more broadly, like a security, and that, of course, drew the attention of the SEC.

What constitutes a security, though? What qualities of The DAO lead the SEC to their conclusion? The SEC spent most of the report elaborating on the answer to this question, and it all stemmed from the following quote:

Under Section 2(a)(1) of the Securities Act and Section 3(a)(10) of the Exchange Act, a security includes “an investment contract.” See 15 U.S.C. §§ 77b-77c. An investment contract is an investment of money in a common enterprise with a reasonable expectation of profits to be derived from the entrepreneurial or managerial efforts of others. See SEC v. Edwards, 540 U.S. 389, 393 (2004); SEC v. W.J. Howey Co., 328 U.S. 293, 301 (1946); see also United Housing Found., Inc. v. Forman, 421 U.S. 837, 852–53 (1975)

The DAO matched this definition of an investment contract, and investment contracts are securities.

Does this apply to every token or coin created? Is every Initial Coin Offering about to incur the wrath of the SEC? In my opinion, no.

Why? Because not every coin offering behaves like stock in a company or other forms of securities as defined by Section 2(a)(1) of the Securities Act and Section 3(a)(10) of the Exchange Act (see 15 U.S.C. §§ 77b-77c, which can be found here). Some do. Many do not, especially the “investment contract” notion used against The DAO.

Many of the tokens and coins being sold in offerings today behave more like Kickstarter campaigns than they do securities. Or gift cards. Or in-game economies like the classic, well-known economy for buying and trading items in Team Fortress 2, a popular game by Valve.

In these cases, you are buying into a presale for a product (Kickstater), a thing that can be exchanged for a product (gift card), or a product itself (Team Fortress 2 items). You are buying a product in some form or another (or potentially a service instead of a product, but I will use the term product to include products and services). You are not buying a reasonable expectation of profits derived from the managerial efforts of others.

And people sell, trade, and profit from products all the time without running afoul of securities law. In fact, in Reves v. Ernst & Young 494 U.S. 56 (1990), a case involving the sale of promissory notes, Justice Marshall wrote:

If the seller’s purpose is to raise money for the general use of a business enterprise or to finance substantial investments and the buyer is interested primarily in the profit the note is expected to generate, the instrument is likely to be a “security.” If the note is exchanged to facilitate the purchase and sale of a minor asset or consumer good, to correct for the seller’s cash-flow difficulties, or to advance some other commercial or consumer purpose, on the other hand, the note is less sensibly described as a “security.”

Take the following scenario: Sally created a business selling widgets and sold gift cards for said widgets. People trade those gift cards in many different ways. It would seem silly to subject Sally (the issuer) to an accreditation requirement for purchasers or to make gift card trading platforms “register as a national securities exchange.” It would seem silly, because Sally’s goal is to sell products, not to deliver investment value to people who bought gift cards.

What about the Kickstarter analogy? Sally can create a campaign on Kickstarter, and as a reward for backing Sally’s campaign, you receive a product at some point in the future. Sally is preselling a product. Presale receipts are not a security.

In terms of the Team Fortress example (in-game economy), the company and community members create digital goods to buy and sell on what essentially amounts to an exchange. None of this is registered as a security, and the exchange is not registered as an exchange, despite the fact that people are spending money, people expect to profit from these transactions, and the ecosystem depends on Valve continuing to support it (remember, “an investment contract is an investment of money in a common enterprise with a reasonable expectation of profits to be derived from the entrepreneurial or managerial efforts of others”).

In all of these cases, no one thinks of these things as securities. Or do they? What if Sally sold gift cards at a discount? What if an expectation of profit formed, because people can flip the gift cards and pocket the difference? This at least seems like a “reasonable expectation of profits.” There could be an argument made that it is “derived from the managerial efforts of others,” too, because if the business fails, the gift cards are worthless. Are the gift cards now a security? Have we found a flaw in this line of thinking?

Conveniently, the American Bar Association published an article titled, Small Business Fundraising Ideas: Avoiding Complex Securities Regulations (PDF). In it, they address both Kickstarter and gift cards with references to case law.

Regarding Kickstarter, they say:

[A] benefit of this financing option is that federal securities laws generally do not apply to this kind of transaction because products, rather than securities, are sold. Forman, 421 U.S. at 852 (1975). State securities regulators, however, may have a different point of view. For example, in 1959 a country club in California presold club memberships and used the membership funds to start the business. Under federal securities laws, these memberships likely would not be considered securities, but a California court found that there was a significant risk that the business might fail before members got a chance to use their membership benefits. Because of the risk of loss involved, the memberships were deemed securities, and the club was found to be in violation of state securities laws for not registering the membership sales with California’s state securities regulator. Silver Hills Country Club v. Sobieski, 55 Cal. 2d 811, 814–16 (1961). But see Moreland v. Dep’t of Corp., 194 Cal. App. 3d 506, 522 (1987).
Roughly 17 states, including California, use the risk-capital test to determine whether a presold good might be considered a security.

In other words, if you might fail before delivering your product, you might be a security. In some sense, this is saying the buyer has a “reasonable expectation of profits” (receiving the product) “derived from the managerial efforts of others” (the company succeeding to deliver said product).

However, we should consider that the value “derived from the managerial efforts of others” is not to the same degree as stock in a company. With Kickstarter, it is a one time thing (you get your product or you do not) versus stock which has an ongoing expectation of value creation. Regardless, this is why it is important for Kickstarter campaigns to have a well thought out minimum so you do not fail.

Now let’s look at what they say about gift cards:

As far as securities laws go, selling regular gift cards is very similar to preselling products. … Discounted gift cards, however, may be cause for concern from a regulator’s perspective. For example, if you sell a gift card with a purchasing value of $20 for only $15, then the gift card owner technically has the expectation of receiving $5 worth of extra value for his or her money. Unfortunately, this looks to regulators fairly similar to hedging futures on Wall Street.
As a general rule of thumb, securities regulators usually look to the substance and characteristics of a transaction, rather than the label you use for it. Sec. & Exch. Comm’n v. Howey Co., 328 U.S. 293, 298 (1946). See also Tcherepnin v. Knight, 389 U.S. 332, 339 (1967). Anything that creates an expectation of an increased value or profit, over and above what was given in exchange for it, can be characterized as a security.

In other words, if you are selling a discounted gift card, and people expect to make a profit from your discount, then we might find a “reasonable expectation of profits.” The managerial aspect applies again, because if you fail, the gift card is worthless. However, as with Kickstarter, failure to stay in business should be avoidable if you manage your cash appropriately and do not overcommit your ability to deliver product to gift card holders.

At the end of the day, lots of companies presell product or gift cards, even at discounts. Lots of companies do so without failing. Lots of people trade products and gift cards. Lots of companies facilitate trades. Lots of these do not cause any action by the SEC.

For example, take people who preordered the Nintendo Switch and then resold it at a profit on Ebay because of scarce supply. Nintendo did not have to register the Switch as a security. The buyers did not have to be accredited. Ebay is not a registered securities exchange. Did the SEC take anyone to court? No.

Take another example: Amazon Web Services (AWS). I can prepay for server time and get a discount. This leads to a higher expectation of profit for whatever service I am running. I can even resell prepaid server time to other AWS users. AWS has to deliver on its promise to provide the server. It might be possible to argue that I am investing money with a reasonable expectation of profit derived from the managerial efforts of others. Despite that, server time is not a registered security. The buyers do not have to be accredited. AWS is is not a registered securities exchange. The SEC is not going after AWS.

One more example: were Beanie Babies securities?

I could keep going, but you see the point. Clearly there is a lot more nuance than meets the eye. Nintendo and Amazon are not at risk of failure, so they pass the risk-capital test, but both are creating a situation where people expect to profit. Beanie Babies certainly created a fervor of profit expectations. Should their products be considered securities?

We could say yes, they are all securities, but that opens Pandora’s box. It would bring into question so many things that seem normal and are not labeled securities today. I believe this is a strong reason regulators are so hesitant on blockchains (as they should be).

Time to bring this back to ICOs. As you probably guessed, I believe many ICOs are essentially not very different from Kickstarter campaigns, gift cards, or in-game economies.

One problem is that it is difficult in some cases to peg a value to a token offered in an ICO. The value of the token is not as obvious as a gift card. Consequently, amateur speculators enter the market, pushing the price of tokens up, causing more profit seekers to enter into the market, and the cycle repeats. People develop an expectation of profit, but is this a “reasonable expectation of profits”? Or is this unreasonable, like Beanie Babies? The company was just trying to presell widgets with no expectation that anyone would profit from widget prices skyrocketing in second hand sales, yet somehow, they have skyrocketed. Consumer demand unreasonably drove an expectation of profit irrespective of actions the company took. Should this be a security?

For example, take Golem, the peer-to-peer CPU rental network with its own token based economy. Golem raised roughly $8.6M worth of Ethereum by preselling Golem Network Tokens (GNT), allowing them to fund development of the network software. GNT will be redeemable for CPU time on the network.

Does GNT have a reasonable expectation of profit? It certainly does not in the sense that you expect profit when you buy stock in a company (you expect the company to return value to shareholders either through market cap or dividends). GNT might return a profit, just as Beanie Babies might have, but like we said before, something is not a security simply because it could increase in value. A rare timepiece could increase in value. It is not a security.

In practice, what is and is not a security is something that has been clarified over years of case law. Many people point to SEC v. W. J. Howey Co., 328 U.S. 293 (1946) as a pivotal case in the matter, establishing what is known as the Howey test. The case centered around whether or not a land sale and service contract was an “investment contract” (i.e., a security). In the opinion, Justice Murphey wrote:

The test is whether the scheme involves an investment of money in a common enterprise with profits to come solely from the efforts of others. If that test be satisfied, it is immaterial whether the enterprise is speculative or nonspeculative, or whether there is a sale of property with or without intrinsic value.

Over time, more cases elaborated on what constitutes a security. For example, in Reves v. Ernst & Young 494 U.S. 56 (1990), the case I cited before, a wider quote provides a more thorough perspective:

First, we examine the transaction to assess the motivations that would prompt a reasonable seller and buyer to enter into it. If the seller’s purpose is to raise money for the general use of a business enterprise or to finance substantial investments and the buyer is interested primarily in the profit the note is expected to generate, the instrument is likely to be a “security.” If the note is exchanged to facilitate the purchase and sale of a minor asset or consumer good, to correct for the seller’s cash-flow difficulties, or to advance some other commercial or consumer purpose, on the other hand, the note is less sensibly described as a “security.” See, e.g., Forman, 421 U.S. at 421 U. S. 851 (share of “stock” carrying a right to subsidized housing not a security because “the inducement to purchase was solely to acquire subsidized low-cost living space; it was not to invest for profit”).
Second, we examine the “plan of distribution” of the instrument, SEC v. C.M. Joiner Leasing Corp., 320 U. S. 344, 320 U. S. 353 (1943), to determine whether it is an instrument in which there is “common trading for speculation or investment,” id. at 320 U. S. 351.
Third, we examine the reasonable expectations of the investing public: The Court will consider instruments to be “securities” on the basis of such public expectations, even where an economic analysis of the circumstances of the particular transaction might suggest that the instruments are not “securities” as used in that transaction.
Finally, we examine whether some factor such as the existence of another regulatory scheme significantly reduces the risk of the instrument, thereby rendering application of the Securities Acts unnecessary.

Justice Marshall boils it down to four tests. The more something matches, the more likely it is security. The tests are essentially as follows:

  1. Is the seller primarily raising money and the buyer primarily interested in profit?
  2. Will there be common trading for speculation or investment of the instrument?
  3. What are the expectations of the public?
  4. Does other regulation exists to reduce the risk of the instrument?

These four tests will be tough for some ICOs if they want to avoid classification as a security. Much of this will come down to positioning and setting expectations.

For example, ICO issuers could influence the expectations of the public by encouraging people to only buy tokens if they have an intent to use the tokens. That said, they cannot necessarily control whether or not speculators will buy the tokens.

However, I do believe these are not insurmountable issues, even if it requires changes in how ICOs go to market and operate. Many of these tests can be applied to things like Kickstarter campaigns or AWS server reservations, and those are not securities. I believe we have some wiggle room.

For example, apply these tests against any in-game economy, and you will discover that securities law is not black and white. The sellers in those economies (which include the company that created the game) are expecting to “raise money for the general use of a business enterprise” (test 1). There is most certainly common trading in game items (test 2). The public expects items will become rare which might drive them to speculate (test 3), and there might not be a ton of regulation out there to help anyone who loses money in a game economy (test 4).

Ultimately, it will be up to issuers and regulators to work this out. It would be wise to do this through cooperation and not act brashly. Issuers that work with regulators will probably be ok. The real wrath will be reserved for the scams, tricksters, and other fraudulent projects simply out of a primary desire to protect investors and consumers.

In some sense, though, it is a good thing if we start treating some ICOs as securities. It would imply an expectation of profit, and that is good for certain people. Being a security can be a good thing. It also puts us on a path to have tokenized ownership of companies instead of everyone just doing app tokens to avoid securities law.

In the meantime, it would be best to take a conservative approach. If you want to do business in the US, then consider taking precautions. Work with people like Coinlist to do it properly.

Wrapping this up, here are key takeaways:

  1. All ICOs are not created equal. The DAO being a security issuance does not make every ICO a security.
  2. There are tests to consider when determining whether or not something is a security.
  3. We need to be careful which ICOs are labeled securities, because if we overreach, other things might be considered securities that were not before.
  4. Take a thoughtful approach to doing ICOs in the US. Work with experienced professionals.

Please do share other perspectives you might have. Thank you for reading!