Not Legal Advice, 18/11/2019 – The federal government needs to stop using the word “decentralization”

Welcome back to another edition of Not Legal Advice

One of the issues with a weekly blog series is that it is easy to get out of the habit of writing it if you miss one or two. Last week I was slammed traveling all over the East Coast so I didn’t have a ton of time to read or write – although running a solo practice is great, arguably the biggest downside is that you can’t delegate – meaning that clawing back time to write can be a challenge.

But dammit, I’m going to do this every week. This week, I have only one thing to talk about. And that’s the fact that the federal government can’t make up its mind about how to deal with Ethereum.

For those of you just joining the conversation, Ethereum is a cryptocurrency that was created following a $20 million initial coin offering, or ICO, in 2014, in which approximately 70% of the outstanding supply of the coin was sold.

Despite the fact that Ethereum was run in exactly the same way as every ICO the SEC has enforced against, in 2018 the SEC punted on regulating Ethereum following an intense lobbying campaign by Coin Center and other industry participants.

This year, the new Chairman of the CFTC Heath Tarbert has said the Ethereum-where-70%-of-the-tokens-were-sold-in-an-ICO-scheme is also not a security but a commodity falling under his jurisdiction. This was only last month.

Now, we are told, the planned migration of the scheme from proof of work to proof of stake will render the scheme, potentially, a security. CFTC Chief Heath Tarbert:

“Mining is, by its very nature, more decentralized as compared to a stake which reduces energy costs by giving it just one validator or a line of validators,” Tarbert said in response to a question whether ether 2.0 will be classified as securities.

This led to the somewhat unsatisfactory position that, in the eyes of the SEC and CFTC, something could start life as a security, transmogrify into a non-security, and then transmogrify back again after a code fork.

Where is the government coming up with this? Probably from the decentralization bros with a lot of money to spend on lobbying, such as Coin Center.

Ethereum bros of course, not to be outdone, swiftly retorted that mining was in fact the more centralized scheme because it costs money to… buy mining equipment?

“In ethereum’s PoS, the capital that you need to acquire to participate is much more readily available. … Converting capital into an asset that allows you to stake in the protocol is much cleaner.”

*silent screaming*

Where to begin with this. To start, there is no agreed-upon definition of “decentralization” anywhere. Anyone who says that there is such a definition has an agenda to push. And to the extent that the term means anything it’s a relative definition rather than a description of an irreversible, apotheotic end-state.

Mining may be accurately described as decentralized, if there are a wide range of validators distributed geographically widely, and there’s something to compare it against. So, e.g., we might say “the Bitcoin network’s mining is more decentralized than the Ripple Network’s validator nodes.” But Ripple might retort by arguing that “more than 50% of the Bitcoin network’s hashppower is controlled by 3 mining pools, whereas our UNL is more widely distributed.” I wouldn’t buy that argument, but it’s one that could be made. Which argument is correct or not depends on what your assumptions are.

Similarly, a network where stake is widely distributed and which requires a 7-out-of-10 vote to append a new block, such as a Tendermint node, might well be more decentralized than either if enough small validators band together to run the network on their own. What if we start out with four thousand stakers and gradually a handful, say one hundred, buy up most of the outstanding stake? Well then it might have fewer validators than Bitcoin and more than Ripple, but more “pools” than Bitcoin and fewer independent operators than Ripple.

All of these statements are correct in their own way. They’re all contradictory in their own way, too. Which indicates, at least to me, that using “decentralization” as a metric to determine whether something is or is not an investment contract – as the much-vaunted Hinman Test seeks to do – really is not a good idea, because nobody knows what the word means. Yet our regulators keep using this word, and coin bros keep encouraging them to, because asking whether a system is truly decentralized or not is like asking how many angels can fit on the head of a pin. Decentralization is a relative measurement, not a platonic ideal.

Now perhaps – and bear with me here – a better measure of determining whether a scheme is or isn’t an investment contract is to look at how the scheme began and, where there’s a premine which is pre-sold, a view is taken as to the nature of the instrument arising therefrom. Schemes like Bitcoin which arise from work and boot-strapping fall outside of the regulatory perimeter and schemes that do otherwise do not.

Such an approach would be consistent with most of the enforcement which has occurred to date. It would not be consistent with the treatment of Ethereum, for indeed nothing is consistent where Ethereum is concerned, and that confuses the hell out of everyone.

The “decentralization” test confuses the market and is not fit for purpose. Nobody knows what “decentralization” means. Even if we did, as I mentioned, decentralization is a relative rather than absolute measure.

Absolute measures create certainty in the marketplace. And although Howey is not absolute, I can think of one absolute measure which would be entirely consistent with the SEC’s enforcement and coin schemes’ compliance attempts to date – entirely consistent, that is, if Ethereum had been brought into the regulatory perimeter instead of being treated as some kind of outlier. Treat presales of coins (not coupons) as dispositive and abandon the idea that something can start life as a security, transmute into a non-security and then, as a consequence of a code fork, revert back again – which is the current view of the regulators.

As I always say when this issue comes up, I know of no precedent for this anywhere in American law and have yet to have anyone point one out to me in six years of asking this question.

And here is a picture of a marmot.


2 thoughts on “Not Legal Advice, 18/11/2019 – The federal government needs to stop using the word “decentralization””

  1. Excellent post, Preston, thanks so much for putting in the effort, applying thoughtful and thought-provoking analysis, and posting it. Nicely done.

Comments are closed.