Get your tickets to the 2024 Coin Center Annual Dinner 

The SAFT is a reasonable approach to securities law when preselling useful network tokens

New open source framework gives entrepreneurs an option for legally compliant token distributions.

Earlier this month, the Filecoin token sale ended, and for many in the cryptocurrency space this one was different. Why? Because Protocol Labs, the principal developers behind IPFS and Filecoin, decided to be very cautious with respect to U.S. securities and financial surveillance laws and, as a consequence, they only pre-sold Filecoin distributions to identified, accredited investors. That means that you could not have participated unless you could prove that you have annual income over $200,000 for the last two years, or have a total net worth over $1 million.

For some, this may have been disappointing. It means that many individuals were unable to give money to a project they had good reason to be excited about and that they wished to support. It means they will have to wait until Filecoin is a running network to obtain Filecoins for use in this future decentralized storage solution, and at that future point Filecoins may be more expensive.

The reason why this was a prudent step for Protocol Labs is that there remain grave uncertainties with respect to how U.S. securities laws and financial surveillance laws apply to the sellers of newly invented tokens and cryptocurrencies.

As we’ve said several times before, we believe there is a good faith argument to be made that sales of a truly innovative and decentralized token, one that provides or will provide real utility to users, does not (a) fit the formal legal definition of an investment contract, and (b) should therefore not be regulated as a security. Just as there are good reasons why the inventor of a smartwatch does not need to register with the SEC before they raise money on Kickstarter, and just as the owner of a Manhattan apartment building need not register before they sell cooperative housing shares, we believe there are good reasons why the inventor of a new and useful protocol token should not need to register.

The SEC’s recent release of an investigative report on the DAO confirms both that (A) the Howey test for an investment contract is the correct test to determine whether a token is a security, and (B) that some profit-sharing tokens are securities. That report does not, however, analyze whether a token designed to provide a consumptive good or service rather than share profits would not be a security. It’s up to innovators to make that argument.

As we’ve also said several times before, there is a good faith argument to be made that these sales would also not qualify as money transmission or running a money services business. That would mean the seller is not required under law to register with FinCEN and develop extensive know-you-customer and suspicious activity reporting policies. We believe this argument is compelling because the seller of an initial distribution of tokens is merely selling something they own, not serving as a financial intermediary or third-party between two persons engaged in a monetary transaction. Just as a person or business buying or selling any consumer good for their own purposes (and not on behalf of others) should not be treated as a financial institution with surveillance obligations, a person or business selling their own property on their own behalf should also be excluded.

But, to be very clear, both of those “good faith arguments” are just that; they are arguments not legal certainties. It will fall to the actual innovators and developers of these technologies to make those arguments in court and win or lose. Losing can mean massive monetary penalties and even criminal liability and jail. We should never criticize a developer for refusing to fight it out in court, for refusing to put their livelihood and liberty on the line to win an argument.

In the meantime, the approach taken by Protocol Labs is commendable. They’ve developed a strategy that takes advantages of certain safe harbors within U.S. securities laws and have found a way to raise money by offering tokens only to accredited investors in advance of the launch of their platform. Once the platform is live, those investors, as well as Protocol Labs itself, should be free to sell Filecoins to the public at large, the speculative security-like aspect of the token having become secondary to their actual use value (as access to commoditized cloud storage) on the network.

Protocol Labs also deserves credit for open-sourcing this legal strategy. At the core of the strategy lies the Simple Agreement for Future Tokens (SAFT). This is the actual contract between a token seller and their accredited investors, and it has been carefully drafted to avoid issues with uncertainty in the application of securities and financial surveillance laws. Future developers, should they want to take a similar and justifiably cautious approach to a sale, are free to borrow this agreement and suit it to their own purposes, and learn about its structure in the accompanying white paper.

Is this now the only legally compliant way to do a token sale with U.S. investors? No.

First, while the SAFT legal structure is a smart and promising way to address the uncertainty in the law, it is still untested. Eventually the SAFT term will end, and—if all goes to plan—the buyers will receive actual tokens that have a consumptive use on the decentralized computing platform. At that point it is still an open question whether those tokens fit the definition of securities here in the U.S., although existing useful tokens like bitcoin and ether are in the same boat and seem, at this point, a poor fit for classification as securities. Only if the resultant tokens don’t qualify as securities upon the termination of the SAFT can they be freely sold to U.S. persons hoping to use the network’s functionality (and regardless of whether they are wealthy and accredited investors).

Second, there are other ways to minimize regulatory risk aside from the SAFT, and it’s premature to suggest that one or the other is a better approach. Other possible models include:

  • A project could simply not pre-sell a token. The network token could emerge exclusively through mining or in return for valuable participation in the decentralized computing platform. This is how Bitcoin, Litecoin, and many others have done it.
  • A project could refrain from making any sales before the network is live and only sell tokens once the computing system is delivering functional utility in return for tokens. At that point the project’s tokens will be appraised with respect to securities laws just as a SAFT project’s tokens will be appraised upon termination of the SAFT. SteemIt, a decentralized social media reddit project, took this approach.
  • A project could incorporate as a normal company and raise money for development by selling equity or promising tokens to a small number of accredited investors, and then when the network is built some share of tokens that emerge on the platform could be allocated to the development company in return for performing that early work and research. This is generally the approach taken by Zcash.

There are many possible alternatives. None are guaranteed to “pass the Howey Test” because that’s not how securities law works unless the relevant regulators issue bright-line guidance about how they intend to apply a flexible test. But we’re glad to see good faith efforts to comply with the law emerge, and we’re happy to support the open source SAFT project in the hopes it is useful to projects who want to fund the development of unowned Internet infrastructure with the presale of useful network tokens.