The SAFT Whitepaper: What Makes a Token “Genuinely Useful”?

TL;DR

Protocol Labs, Inc., the creator of the Filecoin network, and their counsel, Cooley LLP, recently released a framework for conducting compliant token sales using a Simple Agreement for Future Tokens (SAFT) (the “SAFT Whitepaper”). One of the SAFT Whitepaper’s conclusions is that if a token is genuinely useful on a functional network, then it is not a security. Due to the complexities of developing and maintaining tokens, determining whether they are genuinely useful is not as easy as you may believe after reading the SAFT Whitepaper. A comparison of three tokens, Bitcoin, Augur and Filecoin, shows that genuine usefulness varies with use case and specifically, the newness of the use case. Ultimately, the determination will need to be made by developers and their lawyers after a fact-intensive inquiry that is specific to each token.

— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — —

The SAFT Whitepaper covers token sale compliance in the context of securities, money transmission and tax laws, but this post will focus specifically on compliance with securities laws. It is therefore essential to know the test for determining whether something is a security, which is known as the “Howey test,” and how it can be applied to tokens. Under the Howey test, if your token includes (1) an investment of money, (2) in a common enterprise, (3) with the expectation of profit, (4) solely from the effort of others, then your token is a security that must be registered with the SEC or fall under a registration exemption. The SAFT Whitepaper doesn’t contest that tokens almost always satisfy Prongs (1)-(2) and therefore this post will focus on the “expectation of profit” and “effort of others” prongs. As a final bit of background, it is also essential to understand that (i) the “expectation of profit” prong can be outweighed by an expectation to use the token and (ii) the “effort of others” prong should focus on the “essential” effort of others.

Now back to the SAFT Whitepaper.

The SAFT Whitepaper starts its Howey test analysis by drawing a distinction between two types of tokens: security tokens and utility tokens.

Security tokens: These tokens are mere representations of traditional Howey test securities and are unquestionably securities.

Utility tokens: These tokens represent the right to use a decentralized blockchain network and whether they are securities depends on the functionality of the token when they are initially sold. Utility tokens can therefore be further divided into “pre-functional tokens” and “already-functional tokens.”

Pre-functional tokens: These tokens are sold before the token developers have even built the network and are most likely securities because (1) the expectation to profit from the token outweighs the expectation to use the token and (2) the effort of the developers to make the token useful is essential to profiting from the token.

Already-functional tokens: These tokens are sold after the token developers have built the network and whether they are securities depends on whether these tokens are:

genuinely useful such that they are actually used on a functional network

If the token satisfies the above condition then it will fail the “expectation of profit” and/or the “effort of others” prongs because (1) anyone who purchases the token to use it will fail the “expectation of profit” prong because their genuine use on a functional network outweighs the profit motive and (2) anyone who purchases the token for profit will fail the “effort of others” prong because they have a genuinely useful token on a functional network where the “essential” efforts of others have already occurred.

This line of reasoning reminds me of a recent post by Daniel Jefferies about the distinction between certain coins and securities that is worth sharing:

The platform may or may not take off. But here’s the thing, the coin that funded the project may remain, even if the project fails. If people like the properties of the coin, such as the anonymity properties of something like Zcash, then the coin could prosper even if the project and its original founders fade away.

Nothing like that exists today. If you buy stock in a company and that company goes out of business, your stock is worth zero. But your cryptocoin might still hold value long after a project turned to dust and that’s an incredible breakthrough. It means you still may get an ROI even without the company succeeding.

Like the SAFT Whitepaper, Jefferies believes that profiting (e.g., getting a ROI) on a token does not depend on the efforts of others (e.g., founders, company, project) after the token is genuinely useful.[1]

I tend to agree, but I do not think determining whether a token is genuinely useful is as easy as the SAFT Whitepaper or Jeffries would have you believe. In determining usefulness, we must focus in on the actual use case for the token.

Take Bitcoin for instance. I would argue that for most of its existence Bitcoin has been genuinely useful as a store of value. I believe this usefulness stems from the fact that a store of value is not a new concept. After all, humans have been storing value with a wide range of objects for thousands of years. We therefore know what features are needed for a useful store of value and find this functionality in Bitcoin because, among other things, it has limited supply, is secure, transportable, divisible and auditable. There are of course bugs in the code that require updates and I would be remiss not to point out the volatility and “FUD” surrounding hardforks and the scaling debate. But even if the Bitcoin Core developers disappeared and we could no longer update or hardfork Bitcoin, people could still use Bitcoin.

Now if we compare Bitcoin’s store of value function to another token’s, like Augur’s prediction market function, we see that determining usefulness is not always so easy. Predication markets may not be new in theory, but they are certainly new in practice. They have only recently gained practical functions because of smart contracts. This means that despite how useful Augur may seem upon ICO, it is impossible to know what additional features will be needed to make predication markets useful in the future. As a thought experiment, we can imagine what would have happened if Augur’s developers disappeared after the ICO. The code would still work, but when a competitor like Gnosis comes along with new additional (i.e., “essential”) features, Augur must also add these features or else it becomes useless and the value of its tokens goes to zero. Can we therefore say that the efforts of others are no longer essential upon Augur’s ICO? I doubt it.

We can also look to Filecoin, which has a file storage function that I believe lies in between Bitcoin’s store of value function and Augur’s predication market function on the spectrum of newness. In the Filecoin ICO, purchasers signed a SAFT and won’t be given their tokens until the tokens are at least “genuinely useful”. Now suppose that on day-1 of usefulness the Filecoin developers disappear. Filecoin will no doubt be useful with respect to competitors like SIA, Storj, MaidSafeCoin and whatever other storage-based protocols ICO in the meantime, but what happens when a new storage-based innovation comes along and there are no Filecoin developers around to improve the protocol? Like the Augur example, Filecoin will cease to be competitive, will no longer be useful and the value of its tokens will go to zero.

After going through the Augur and Filecoin examples one could even look at Bitcoin and change their conclusion to find that the efforts of developers are, in fact, essential, and therefore Bitcoin is not yet genuinely useful. After all, if Bitcoin hadn’t implemented Segwit then another token like Litecoin might have replaced it. This is evident from the rally in Litecoin’s price and the volatility in Bitcoin’s price leading up to Litecoin’s activation of Segwit. So again, we see that developers are essential to a token’s price even after it is apparently useful.

None of this is to say that Bitcoin, Augur or Filecoin should be securities. This is more an exercise to demonstrate that proving genuine usefulness for tokens is complex.Determining genuine usefulness is hard because tokens are not simply the currencies we imagine them to be, they are something way weirder. Applying a traditional analysis like the “effort of others” prong to tokens is, as the saying goes, like trying to fit a square peg in a round hole. Maybe this is because there are so many other “others” when it comes to tokens. We’ve discussed the efforts of developers, but we haven’t even mentioned miners or stakers, who are also essential to maintaining the security of proof-of-work (PoW) and proof-of-stake (PoS) tokens, respectively. People also often overlook the fact that tokens and blockchains are as much a social innovation as they are a technological one, and an innovation that includes the disciplines of computer science, cryptography, economics and game theory all wrapped into one.

If the SAFT becomes a widely-used framework for conducting compliant ICOs, then the genuine usefulness of a token will be a conclusion that developers and their lawyers make after a fact-intensive inquiry that is specific to each token. There is not necessarily anything wrong with this, and it is in fact the same conclusion that the SEC has reached.

I believe the SAFT Whitepaper is an important step forward in unleashing the potential inherent in tokens and blockchain technology. But it is important to note that the SAFT Whitepaper seeks to fit within existing regulation, and with the complexities surrounding tokens and blockchain technology, we may need an entirely new paradigm and legislative reform to properly regulate and innovate in this space.

— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — —

Special thank you to Michael Katz and Adam Salm for reviewing and providing useful feedback.

[1] This is also a good policy argument for why already-functional tokens are not as dangerous to retail investors as securities.