The Rube Goldberg Machine: Security Tokens and the Indirect Holding System
The takeaway: security ownership in the United States is a convoluted mess consisting of multiple layers of intermediate ownership that stems from an archaic system design. Think of the New York subways — the heavy lifting involved in a full revamp of the base will never be preferred to a facelift on the existing structure. Repeated facelifts can lead to absurd results when aggregated together in the face of technological advancement. The SEC rules and interpretations adopted in response to this incumbent ownership system lead to interesting results when analyzed in conjunction with the holder concerns highlighted in my previous article.
Most of you will be surprised to know that you don’t actually own any of the stock in your brokerage accounts. Instead, all of the stock you think is yours is in fact owned by a company you’ve never heard of — Cede & Co. All of the ownership rights you think you possess are an illusion, and what you’re relying on when you receive a dividend or submit a shareholder vote are contractual rights with your broker-dealer (who in turn relies on contractual rights with a company called The Depository Trust Company, whose nominee, Cede & Co., actually possesses those equity entitlements). Surprise!
The vast majority of think pieces I’ve found that touch on the above end with a conspiracy theory about shadow systems designed to control and manipulate the masses…that is certainly not where we’re going here. Instead, I’d like to focus on the ability of digitized financial asset ownership to revamp a dense and convoluted system, bringing order where there is chaos.
A number of leaders in our field, particularly Patrick Byrne at Overstock, have done a great job shedding light on indirect securities ownership and its development in the United States, but it remains one of the least understood aspects of our financial system. Equity ownership in the United States has moved from a system where the issuer and the ultimate shareholders are directly connected to a system in which the beneficial owners of a company are thrice-removed (at best) from physical ownership of any share certificates and, ultimately, not counted as “holders” for purposes of securities laws. In evaluating the merits of this existing system and identifying areas in which blockchain technology might provide for improvements, it is incumbent upon us to take a long look at how the current structure developed, the rationale behind the existing rules and their applicability in a new paradigm. This article is an attempt to lay the groundwork for that evaluation and present some of the questions we will be forced to answer moving forward.
Token legal disclaimer…
The information contained in this article is provided for informational purposes only and should not be construed as legal advice on any subject matter. You should not act or refrain from acting on the basis of any content included in this article without seeking legal or other professional advice.
The Threshold Question
My last article outlined the current regulatory regime in which security tokens must operate and introduced what I termed the “12(g) problem” — the central conflict between the fully democratized ownership aspirations of crypto proponents and the policy objectives of the SEC. The quick and dirty version is that SEC rules provide for hard equity investor limits (500 non-accredited investors and 2,000 total investors) that, if crossed, force companies to begin reporting against their will (this is why Google and Facebook both went public). The threshold question my analysis skipped over is how to actually count holders of securities for purposes of complying with the regulations. This question is not nearly as simple as it seems, and fully explaining the answer will take us down a rabbit hole of contract law theory, history and the intersection of regulation and pragmatism.
“An issuer is not required to register a class of equity securities pursuant to section 12(g)(1) of the Act if on the last day of its most recent fiscal year…the class of equity securities was held of record by fewer than 2,000 persons and fewer than 500 of those persons were not accredited investors.” Emphasis added.
Above is the language of the actual rule that generates the 12(g) problem, and the question is how the SEC calculates the number of securities “held of record” for purposes of determining compliance. SEC rules provide guidance on this definition — paraphrasing 12(g)(5)-1, “for the purpose of determining whether an issuer [has to become a reporting entity], securities shall be deemed to be ‘held of record’ by each person who is identified as the owner of such securities on records of security holders maintained by or on behalf of the issuer, subject to [certain exemptions].” In plain English, the SEC puts the onus on the company or its registrar to maintain a record of security holders, and that record is the source of truth in determining holder counts for purposes of complying with SEC regulations.
This seems easy enough on its surface (not to mention a great example of a centralized ledger for which entrenched third parties are able to extract rents), but let’s take a look at an actual company. Here is a link to Item 5 in Facebook’s most recently filed 10-K. As of 12/31/17, Facebook had 3,967 stockholders of record for its Class A common stock and 52 stockholders of record for its Class B common stock. This should be shocking. How is it possible that an entity with a market capitalization of over $500 billion and that trades over 35 million times daily has a shareholder count in the mere thousands? To answer that question, we need to take a deep dive into the development of immobilized proxy security ownership in the latter half of the 20th century.
“Anyone who has ever dealt with the tax authorities, the educational system or any other complex bureaucracy knows that the truth hardly matters. What’s written on your form is far more important.” — Yuval Noah Harari, Homo Deus
Sweeping statements about the law invite high criticism, but I’ll start from the position that both property and contract law have developed in such a way as to have the transfer of intangible assets mimic the transfer of tangible assets. Harari in both Sapiens and Homo Deus repeatedly promotes human beings’ singular ability to hold abstract concepts in the mind’s eye and coordinate accordingly as the defining feature that led to homo sapiens’ dominating position on Earth. Our entire legal system, nation-states, money and a whole host of other everyday concepts are not “real” in the sense that they exist naturally — they are all fictions a society agrees to believe in as a tool to promote coordination of action. Similarly, the concept of corporate ownership (and of corporations themselves) is merely a legal fiction we all agree to believe in for ease of assigning rights and responsibilities among the various constituencies in a common enterprise.
The legal regime developed to treat shifts in these intangible rights and responsibilities as if they were tangible goods passing between parties. Intangible rights stemming from corporate ownership were legitimized in the form of physical certificates, and the courts endorsed the primacy of physical ownership in determining outcomes in legal disputes. The legitimate transfer of a security certificate from one party to another consisted of a good faith negotiation, an indorsement on the underlying security and, most importantly, physical delivery of the certificate in exchange for value. In the case of bearer instruments, this physical “holding” of the certificate is all that is required in order to redeem the instrument for coupon payments, principal repayments and the like. In the case of registered instruments, an additional step is required to formally complete the transfer of the security — altering the issuer’s registry in order to recognize the new owner.
This additional step is important because it means that “physical custody of a securities certificate does not, in and of itself, give the holder the right against the issuer to enforce the voting, dividend or other rights comprising the security.” (Jeanne L. Schroeder, Is Article 8 Finally Ready This Time? The Radical Reform of Secured Lending on Wall Street; quoting from Prof. Schroeder’s article as I cannot say this line better). Instead, the new holder (or his/her agent — e.g., a broker) would present a physical certificate, properly indorsed, and the issuing company (or its agent — e.g., a broker or a transfer agent) would cancel the security and issue a new one. Only then, once the certificate is physically possessed, is the issuer legally obligated to recognize the new security holder’s claim.
I just summarized centuries worth of case law and security ownership development in two paragraphs, so there is a great deal of nuance I am skipping over, but the above is largely how the securities settlement system worked up until the late 1960s. With the 1960s came a huge market boom and concomitant increase in trading volumes. As trading volume spiked, the back-office clerks at the various brokerages found themselves unable to keep up with the necessary paperwork to effectuate the legal transfer of title to certificated shares. Here is a chronology prepared by the NYSE for the Subcommittee on Commerce and Finance in 1971. The NYSE limited trading hours and even reduced the number of trading days in a week to four in order to ease the burden on the brokerage’s back offices to comply with settlement obligations. About a hundred brokerages failed because of these delivery requirements, and others fell so far behind in their settlement process that “unperformed obligations could range from 70% to 200% of a firm’s total assets.” This operational crisis led to a Congressional investigation and, ultimately, a watershed alteration in how changes in security ownership were documented.
Modern Securities Settlement and Clearance
Two competing solutions emerged in the aftermath of the “paper crunch” in the late 1960s: (1) immobilization and (2) dematerialization. Both solutions would have the effect of vastly reducing the amount of paperwork required to settle security trades, but they approached the problem from opposite sides of the spectrum — the former as a centralized solution consolidating security ownership in a system of intermediaries and the latter as a decentralized solution maintaining a direct relationship between shareholders and the issuer itself. Note — David Donald’s article (linked here again) provides a much more detailed discussion on the creation of the indirect holding system and I highly recommend it for anyone who would like to learn more.
I’ll start with what ended up as the losing proposal — dematerialization. Dematerialization provided for the elimination of certificated shares altogether in favor of book-entry systems held on the register of an individual issuer or its transfer agent. This solution is elegant in that electronic delivery obviates the need for physical delivery, but it faced a number of technical and pragmatic obstacles given the era in which it was proposed. The requisite technology to effectively pursue dematerialization had not been adopted on a large enough scale to make this solution plausible. Further, individual state legislatures would have to accept dematerialized shares as a valid form of share ownership in order for the plan to function on a nationwide basis. This is pure speculation, but perhaps certain entrenched parties maintained economic interests as intermediaries and possessed the political power to ensure any “paper crunch” solution did not remove their role in securities settlement as well.
On the other hand, immobilization took a top-down approach to the paper crunch — if the problem is that the movement of securities triggers onerous paperwork, the solution is to halt the movement of securities. Firms would bifurcate the physical ownership of the securities and beneficial ownership of the intangible rights those securities represent. Thus, all physical certificates were to be immobilized and held in a central depository and brokerages would enable book-entry trading of rights stemming from those underlying securities rather than exchanging the securities themselves. What’s more, brokerages would “net” positions at the end of the day rather than following the linear process of completing full delivery prior to moving to the next transaction. This immobilization approach was adopted into law as part of the 1975 Securities Act Amendments, requiring the SEC to create rules that provided for the placement of securities in a central depository. Two entities were formed in order to serve these holding and netting functions — the Depository Trust Company (“DTC”) to act as a holding agent and the National Securities Clearing Corporation (“NSCC”) to act as a netting agent. These two entities are now housed under the Depositary Trust & Clearing Corporation (“DTCC”), and they perform essentially all of the public market settlement and clearing functions in the United States today.
There is one more player to introduce to avoid confusion — Cede & Company (“Cede & Co.”). Cede & Co. acts as DTC’s nominee as registered holder for all securities deposited into DTC’s vault. In practice, this means Cede & Co. is the technical owner for the vast majority of shares in the United States. Let me explain how.
There are a number of articles and videos explaining the system of intermediaries along the chain of share ownership, but I found this one to be the most fulsome explanation for a layreader. Here is my consolidated version with a real-world example.
- When a company goes public, it will typically engage a lead underwriter to help navigate them through the process of issuing shares to the public market — that lead underwriter is typically given “left” placement on all marketing materials (see Morgan Stanley listed to the left in big letters on the prospectus cover).
- The underwriting agreement is primarily a diligence document for the underwriters to avoid liability for material misstatements or omissions in the offering documents, but it also governs the initial sale of securities to the underwriters prior to those underwriters selling securities to the public. Check out Section 2(e) — all shares sold to the underwriters must be placed with Cede & Co. as nominee of DTC, and those shares must be registered in the name of Cede & Co. on the Company’s share registry (note — technically this is simply a representation of the company, or the selling stockholders in this case, that they will do so, not a covenant, but that nuance is unimportant for purposes of this explanation).
- Finally, take a look at the exhibit — this is the form any certificated shares of DocuSign must take. You’ll see in the fine print that the share is only transferable upon surrender of the certificate properly endorsed as indicated on page two. In practice, these certificated shares are usually printed only in limited amounts to serve as mementos for those involved in the IPO (e.g., they are put in frames on people’s walls), but you can see the “paper crunch” problem still very much exists where certificated shares are issued to individual persons!
In light of everything we’ve discussed above, where do DocuSign’s shares actually reside and how do we keep track of ownership rights and responsibilities? The answer is “simple” — there is likely a global share certificate that represents all of the shares of DocuSign common stock (less mementos). That share physically resides with Cede & Co. as nominee of DTC. DTC, in turn, maintains a book-entry system that keeps track of all DTC members’ (broker-dealers) positions. Those broker-dealers, in turn, maintain book-entry positions for their clients’ accounts. Thus, what we would commonly refer to as a “shareholder” is, at a minimum, three levels removed from actual physical share ownership!
When DocuSign issues a dividend or conducts a shareholder vote, its bylaws and the legal regime generally provide that it only needs to focus on the holders listed in its books — specifically, DocuSign is not “bound to recognize any equitable or other claim to or interest in such share or shares on the part of any other person whether or not it shall have express or other notice thereof.” DocuSign gives notice and payments to Cede & Co., who then passes those payments and notices through DTC to the Broker-Dealers who then ultimately convey those items to the owners on their books. When things go wrong, the results can be…hilarious?
Let’s return to the original problem. Section 12(g) limits the number of equity holders of record before a company is forced to register to 500 unaccredited and 2,000 total investors. We saw with Facebook that its total number of holders is in the mere thousands despite a market cap and trading volumes that would imply a much larger investor base. What we’ve laid out above would indicate that Facebook’s holder of record should be even smaller (that is, one — Cede & Co.), so what are we missing?
Take a look at the SEC’s guidance here (see Section 152). Institutional custodians and other commercial depositories (e.g., Cede & Co.) are specifically excluded from being counted as “single holders of record for purposes of the Exchange Act’s registration and periodic reporting provisions. Instead, each of the depository’s accounts for which the securities are held is a single record holder.” The guidance clarifies that securities held in street name by a broker-dealer are held of record under the rule only by the broker-dealer. From a policy perspective, it is hard to justify the placement of this line — the SEC will “look through” the depository to see the broker-dealers as holders of record, but will not “look through” the broker-dealers to see the ultimate holders. As detailed in my prior article, the whole justification for imposing investor limitations is to ensure that mom- and pop-investors receive adequate disclosure. If that is true, then why refuse to see how many mom- and pop-investors have actual skin in the game for a particular issuer?
The answer is pragmatism. The regulators recognize their own limitations and look to bright-line rules to provide goal posts that will capture the majority of issuers. Indeed, other rules permit them to “look through” to the beneficial owners rather than the holders of record where abuse is suspected (see Rule 12g5–1(b)(3)). When I propose an ideal regulatory environment later on, we’ll take a much closer look at the imposition of these investor thresholds and why “adequate disclosure” may be a wholly inadequate explanation for requiring registration.
Our existing securities ownership system consists of layers of intermediaries, each of whom extracts profits by increasing the costs of transactions. Application of blockchain technology presents the opportunity to re-think this paradigm. Do we still need physical share certificates or does technology now allow us to “dematerialize” intangible ownership? And what about central depositories? With a distributed ledger, shareholders can once again be directly linked to the companies they invest in, but it is fair to ask whether that is the ideal solution. There are real economic gains from specialization — issuers are not experts in counting votes or distributing dividends, but a third-party custodian could be. Determining which intermediaries are necessary and which are rent-seeking incumbents whose entrenched position adds little economic value is a weeding-out that is necessary. A free and open market is the most effective tool for accomplishing this process, but the question remains whether the regulatory changes we implement will allow that market to function appropriately.
Coming soon on this channel: Considering a New Regulatory Framework, Tokenized Airdrops, What is an Equity Security Anyway? and much more…
In writing this post, I drew on an amazingly large number of articles, books and posts. Along with the documents referenced in the article, here is the short list I used for de-mystifying our current securities ownership system:
The Financial Stability Oversight Council (FSOC) released its annual report on the health of the U.S. financial system…caitlin-long.com