cryptocurrency securities

Can the SEC Measure Whether Cryptocurrencies like Bitcoin Are Securities?

by William Restis

published September 7, 2017

I’ve been studying cryptocurrencies and blockchain with considerable interest for some time now.  As a lawyer, one of my favorite intellectual exercises is comparing new phenomena to existing legal standards.


While the world continues to incorporate and adapt to the emergence of blockchain technologies, there is much regulatory uncertainty in the United States. The biggest uncertainty is whether the federal securities laws apply.

A significant portion of the legal community speculate that many existing blockchain assets (“BCA”), and their fundraising methods, seem to fall within these securities laws. If this is correct, then many BCA issuers and exchanges could find themselves in hot water. To put it mildly, a lot of regulations are being ignored. But the “securities” laws only apply if a BCA can satisfy the legal definition of a “security.” Seems simple enough. But since the question hasn’t been tested in the courts, we attorneys are left to our speculation. 

Recently the Securities and Exchange Commission issued an opinion that helps fill in the blanks. The SEC analyzed the Decentralized Autonomous Organization or “DAO,” one of the very first initial coin offerings (“ICO”). The DAO ICO involved the sale of crypto-tokens that acted like shares in a decentralized enterprise. While there are so many interesting issues regarding the DAO, I’ll leave the reader with the SEC’s opinion itself, and many other sources of commentary that have popped up around the Internet. Instead I want to talk about a particular element of the SEC’s analysis.


To qualify as a security, any expectation of profits from a BCA must be “derived from the managerial efforts of others.” SEC Ord at p 12, citing SEC v. Glenn W Turner Enters., Inc. The SEC stated that ”[t]he central issue is ‘whether the efforts made by those other than the investor are the undeniably significant ones, those essential managerial efforts which affect the failure or success of the enterprise.’” 474 F.2D 476, 482 (9th Cir. 1973). In the DAO case, the SEC ultimately determined that certain key actors “were essential to the success of the enterprise.” This helped confirm to the SEC that DAO tokens were in fact securities.

Ok fair enough. Obviously different facts in any given case will move that needle up or down on this issue. But how are we to quantify the “essential[ness]” or “significance” of someone’s managerial efforts?  What if someone wants to establish that a particular BCA fundraising operation is, or is not a sale of securities? In other words, on what sort of evidence can they rely?


Some research recently discussed at the 2017 Blockstack Summit may be useful. Balaji S. Srinivasan gave a talk addressing how decentralization is essential to the health of any blockchain ecosystem. He noted how everyone simply assumes that the more autonomous and decentralized an ecosystem is, the less likely it can be captured by any individual or group within it. But prior to Srinivasan’s research, no one had yet come up with a metric to quantify decentralization in any given BCA.

Srinivasan pointed to a non-blockchain analogy highlighting the decentralization issues: takeover attempts of corporations. For example, a publicly traded corporation can have a market cap so large that no individual or group owns enough shares to take over the company. In such a situation, it would be hard to organize a group of people to get more than 51% of the shares necessary for a takeover.

Alternatively, a closely held corporation is far easier to take over because fewer shareholders are needed to gain a majority. Blockchain ecosystems operate under similar dynamics. Small, closed BCA ecosystems can essentially be hijacked by a small group. Srinivasan argued that being able to measure the susceptibility of a BCA ecosystem to capture, will allow proactive steps to be taken to redistribute power imbalances in real time.

But how are we to quantify decentralization of a blockchain? Srinivasan applied econometric analysis to bitcoin and ether using “Gini” and “Nakamoto” coefficients. Gini coefficients are traditionally used by economists to measure wealth inequality in a society.


The closer a Gini coefficient is to zero, the closer the ecosystem is to perfect equality. An example would be a network with 100 nodes, with each node getting one vote in the system. A Gini coefficient of one would mean a single individual controlled all the nodes. Applying Gini coefficients to various metrics of a blockchain ecosystem can thus help asses the relative decentralization of any component within it. For example, we can get a Gini coefficient related to miner block rewards. Mining hash power might be evenly distributed (low coefficient) or highly concentrated (high Gini coefficient). Exchange volume, node distribution, and even ownership of tokens can be measured. This is because BCA ecosystems are all defined by precise mathematical relationships calculated in real-time. Gini coefficients can paint a picture of BCA decentralization.

Nakamoto coefficients provide additional context and analysis. A Nakamoto coefficient measures the number of individual participants within an ecosystem whose participation is necessary to pass the threshold of control. Let’s go back to our corporation example. If we have 10 shareholders, each with a different percentage of shares, what is the minimum number of shareholders necessary to gain a controlling 51% interest? The lower number of participants necessary to gain control, the easier it is to take over the BCA.


Why am I writing about all this? What does the above have to do with proving whether or not a particular BCA is a “security”? In many ways Srinivasan’s concern for decentralization is a proxy for the ability of key individuals to control a BCA. In other words, Srinivasan’s analysis not only measures whether a BCA can be controlled, but also whether it is in fact controlled by key individuals or groups.

The SEC spent considerable effort analyzing the development, operation, maintenance, and ownership of the DAO. This seems to be a key part of the SEC’s attempt to determine whether profit expectations from the DAO were “derived from the managerial efforts of others.” Srinivasan’s various metrics allow many features of a BCA ecosystem to be examined empirically, and accordingly weighed.


So how do bitcoin and ether stack up? The results of Srinivasan’s analysis follows:

Now I’m no mathematician, and we’ll have to leave it to the experts to argue the weight of the evidence… but my anecdotal reaction for both bitcoin and ether is that the Gini coefficients seem high. And the Nakamoto coefficients seem low.  The analysis indicates that for most available metrics, ownership and control is fairly concentrated, with only a few individuals necessary to gain control of key features of the bitcoin and ether networks.

While I personally disagree with this conclusion, Srinivasan’s analysis weighs in favor of bitcoin and ether being labeled “securities” under US law. But we won’t know until the issue plays itself out in the courts.


I am not your attorney. This is not legal advice and you should consult your own counsel.

Restis Law Firm Signature Facebook icon Twitter icon Instagram icon Pinterest icon Google+ icon YouTube icon LinkedIn icon Contact icon