The Security Token Thesis

Last summer I wrote , in which I hypothesized that a broad array of assets will move to blockchain records of ownership (represented by tokens), thereby changing the way society holds and transfers investments. A lot has transpired since last summer. In short, it’s happening — the infrastructure to support security tokens is now being built out. Traditional Asset Tokenization I define security tokens as any blockchain based representation of value that is subject to regulation under security laws. That includes tokens representing traditional assets like equity, debt, derivatives, and real estate, and it also includes pre-launch utility tokens that are deemed securities by the SEC. In this article, I focus on the traditional asset segment and lay out the reasons why security tokens dominate other methods of recording and trading ownership claims. Taken together, the features listed below form the foundation of the thesis that security tokens will see widespread adoption across numerous asset classes in the coming years. 24/7 markets Fractional ownership Rapid settlement Reduction in direct costs Increased liquidity and market depth Automated compliance Asset interoperability Expansion of the design space for security contracts The first four are pretty straight-forward, and I’ve already written about liquidity and automated compliance in other posts, so I’ll review these six topics only briefly. I will state up front that some of these features could be delivered through a relational database, and yet, they haven’t been. Why not? I offer some initial thoughts in the section. I conclude with a section on the potential for innovation in security design, which is the most interesting part of the whole thesis. Interoperability 24/7 Markets The major U.S. stock market exchanges open at 9:30 a.m. and close at 4:00 p.m. Eastern Time each weekday, with the exception of holidays. After the closing bell on Friday, you can’t trade your stock again on major exchanges for over 65 hours. It’s 89 hours on a holiday weekend. Any number of information events can and do happen during these breaks. In fact, companies often after 4:00 p.m. purposely release information However, focusing on the U.S. stock market is too narrow because this is not solely a U.S. phenomenon. Teams all over the world are working on tokenizing assets and some observers suggest that this will accelerate the integration of global markets. Around-the-clock trading hours will accommodate all time zones. It is worth noting here that expanding trading hours comes with a cost. Prior research has found that trading outside regular hours is characterized by . lower liquidity and higher spreads Do we need a for 24/7 trading? Of course not, but it’s part of the package. The blockchain ecosystem is a technology stack where “always open” is the de facto standard for cryptocurrency exchanges, time will tell whether security token exchanges adopt similar policies. blockchain Fractional ownership Fractional ownership is not unique to blockchain, in fact, it’s not even unique to this century. Joint ownership dates back to the Roman Republic, or the Dutch East India Company in more modern times. However, some assets classes such as commercial real estate and fine art continue to be characterized by high unit costs. A typical retail investor cannot harness the resources required to buy a Manhattan high rise. The investor is left with two options: (1) Forego exposure to Manhattan commercial real estate in their investment portfolio, or (2) gain exposure through an intermediary, for example a publicly traded Real Estate Investment Trust (REIT), where it is often bundled with a portfolio of other buildings of varying quality and characteristics. Security tokens offer an efficient path to fractionalize a single high value asset. As more assets are fractionalized, we can achieve more optimal asset allocation at the retail level. It moves us closer to being able to construct a true “ .” market portfolio At scale, this also opens up new investment strategies. Long-short strategies have been used in stock markets for years and these will expand to any asset class with fractionalized ownership. Imagine being able to go long Brooklyn and short Manhattan, thereby creating an “NYC market neutral” real asset portfolio. Admittedly, these strategies require that either a security token lending market or synthetic derivative product develops to facilitate the short. With increased trading activity of fractional ownership, price discovery will be enhanced and markets will become more efficient for assets that have historically traded infrequently due to high unit costs. Rapid settlement The first distinction to understand is the difference between execution and settlement. Trade execution means recording an agreement between a buyer and a seller to exchange a specific quantity of an asset at a specific price. Settlement means that the documentation around the transfer of ownership from the seller to the buyer (and payment from buyer to seller) is recorded and complete. Settlement is when ownership of the assets actually changes hands. Exchanges like NASDAQ and NYSE can trades very quickly, but asset transfers takes time. The SEC a shortened settlement cycle for most broker-dealer transactions to T+2. The “2” is days, meaning the ownership doesn’t change hands for two days after trade execution. Settling transfers of LP and LLC interests can take even longer. execute settling recently adopted A figure from Richard Brown’s on securities settlement illustrates the complexity of settling a public equity trade: 2014 article Trades for bitcoin or ether settle in minutes, not days, but there are a lot more parties involved in securities transactions — more than most investors appreciate. There are complexities such as short selling and margin buying. Blockchain has the potential to increase settlement speed for securities, but it’s more complicated than a comparison to cryptocurrencies. The degree to which these processes can be automated through interoperable smart contracts will determine the gains in settlement speed. Cost Reduction Advisory services around securities issuance are costly, and that’s not going to change in the near term. However, several parts of the issuance process will eventually be automated, which will reduce costs in the long run. Post-issuance, there are a lot of administrative costs around ownership reconciliation. When startups are acquired, reconciling the capitalization table to the underlying purchase agreements and option grants is costly. This problem is exacerbated as companies grow. The story of is instructive: A 2015 court ruling required a payment to all shareholders. There were 36M shares outstanding, but claims for payments were filed for 49M shares. Something clearly doesn’t add up. The resolution boiled down to a reconciliation exercise to figure out who owned what and when. The article linked above dives deeper into what aspects of this issue a blockchain can and can’t solve, but the conclusion is that the current system is “starting to show its age.” Dole Foods When all the ownership claims are tokenized, cap tables will be reconciled in real time by code. All the contractual features such as liquidation preferences, ratchets, and drag-along rights will be baked in to the securities allowing managers to easily run scenario analysis to calculate payoffs under different assumptions. Reduction of direct costs is not the most interesting benefit of security tokens, but if it mitigates enough administrative costs, this alone may be a sufficient condition to induce share-to-token conversions such as the one is currently executing. Anexio Liquidity and Market Depth Most private assets are relatively illiquid, which means the ownership interests are costly to trade. For private assets like an LP interest in a venture capital or private equity fund, exiting the position before fund liquidation frequently involves deep discounts, and often require GP approval. Harbor CEO, Josh Stein, stated the benefit of tokenization succinctly: “lock in the capital without locking in the investors.” Tokenized funds allow the fund managers to invest in illiquid assets without fear of redemptions, while fund investors can access liquidity in the secondary market. A deeper market for the ownership interests and the increase in investor liquidity is expected to be accompanied by an increase in value. This is what economists call a . Beyond VC, private securities of all types are often highly illiquid. Security tokens promise similar gains in liquidity for asset classes like real estate and early stage equity. liquidity premium Divisibility of high unit costs places these assets within reach of a much broader market, but market depth will also be increased through a few other channels: (1) The rise of cryptoasset prices have created billions in crypto wealth, some of which would love to diversify into more stable assets without returning to fiat. (2) Algorithmic market makers such as Bancor show some promise of increasing market depth. (3) Security tokens could mitigate , making it easier for for buyers in one country to access assets in another country. As I discuss in the next section, security tokens allow many compliance features to be automated, potentially relaxing some of the regulatory frictions and facilitating integrated global markets. market segmentation Automated compliance I addressed this topic when came out of stealth in February, but will briefly summarize it here for completeness. (Disclosure: I’m an advisor at Harbor) Harbor The case for tokenized securities often revolves around relaxing frictions to trade, and one of the most complex frictions is adhering to regulations. It is complex for at least two reasons: (1) Regulations can vary along multiple dimensions such as asset type, investor type, buyer jurisdiction, seller jurisdiction, and issuer jurisdiction. Each of these dimensions has numerous regulatory permutations and multiple regulatory agencies that govern trade. (2) Regulatory compliance is typically documented through a series of separate ledgers, each constructed by entities that facilitate issuance and/or secondary market trade. It is only through reconciliation of these ledgers that ownership and compliance is legally validated. In this environment, maintaining compliance adds latency and cost to trade, segments markets, and reduces liquidity. A key feature of security tokens is that they are programmable. Many elements of the contracting environment can be hardwired into the architecture of the security. When securities are tokenized, compliance can be automated, which means that regulated trade will no longer be restricted to walled gardens. Security tokens will be able to trade anywhere, including decentralized exchanges. Further, baking compliance into the token could help market participants navigate the task of selling securities across borders. extremely complex Security tokens may make compliance so frictionless that regulators begin securities to tokenize, an idea I first heard from Anthony Pompliano. This is not as far fetched as it might seem — there is precedent for mandates from the SEC for technology adoption. As far back as 1996, the SEC required of financial statements through EDGAR, and later adopted XML technology. Recall that public adoption of the Internet was only just beginning in 1996, so the SEC was ahead of the curve in this respect. There were only on the Internet in 1996, or 0.9% of the world population. This is not far off the adoption rate of blockchain today. Coinbase alone has over 20 million accounts as of this writing. requiring electronic filing 36 million users Asset Interoperability Interoperability is one of the most important concepts in technology. The Internet itself is essentially a stack of protocols that enable many different types of software to exchange and make use of information (i.e. TCP/IP, SMTP, FTP, SSH, HTTP). It’s the reason I can use Outlook to compose an email, send it from my .edu address to a friend with a .com address, who then reads the email through Google Chrome. The arc of technological evolution bends towards interoperability and interoperability is facilitated by standards. By definition, standards must have widespread buy-in to be effective, which makes it more challenging for centralized solutions to induce market adoption of standards. Blockchain offers us protocol standards upon which everyone can build, and that is a big part of why this is the right technology to re-engineer the financial plumbing today. It will almost certainly not be the right technology forever. Either blockchains will evolve to become more scalable, less resource intensive, and more interoperable, or they will be replaced by new technology with these features. Regardless of what we call it, the important thing is that the “consensus system” remains, providing a decentralized way for (rational) economic actors to reconcile the truth. The economic incentives that induce actors to perform this function without a central coordinator is the true innovation. (h/t to Chris Burniske for this passage) Let’s take a moment to address the question every blockchain skeptic loves to ask: “Do we need a blockchain for this? Couldn’t this be done with a database?” The answer is yes, some of it be done with a centralized database, but it begs the question “why hasn’t it already been done?” The answer is that the current centralized solutions for electronic value transfer lack compatibility — they don’t talk to each other. I can’t send value from PayPal to Venmo, or from E*Trade to RealtyShares. These layers aren’t interoperable. could Interoperability within the Ethereum protocol is facilitated by the ERC-20 token standard, which allows a wallet to hold any token that adheres to the standard. It makes distributions easy. For example, let’s say I own some ERC-20 tokens that represent ownership in an apartment building. Each month the lease payments from the tenants are converted to a ERC-20 stablecoin by the building manager and pushed to all the owners’ wallets in the corresponding proportions. Regardless of which ERC-20 wallet the owner uses, it can hold both the ownership and distribution tokens. The idea of standardized ownership claims is not new, it’s the reason your brokerage can hold shares of stock in many different companies. The innovation is that ERC-20 is an example of a standard that allows me to hold security tokens that represent many types of assets in the same wallet. This is not to say that or that Ethereum is the only protocol that will have these features. Scalability remains a major concern. If cross-chain interoperability becomes a reality, we’ll likely be able to move beyond ERC-20. Many teams are . different ERC-20 is the only solution currently working on this Some ownership claims, like the deed to my house, are literally on paper, but the point is bigger than digital versus paper. Most asset ownership is already represented digitally. I don’t have any paper certificates for the public equities I own, and frankly neither does my broker, although they could produce them if requested. I don’t have any paper for many of the LLCs and Limited Partnerships of which I am a member. These ownership claims were documented on PDFs and signed electronically. The issue is that even though all these claims are represented digitally, the digital systems don’t all play with each other, which inhibits compliant trade. The thesis underpinning the idea that everything will be tokenized is grounded in the aspiration that everything will be interoperable. If the ecosystem for global assets becomes interoperable, it means we can hold ownership claims to a commercial building, early stage equity, corporate bonds, a T-bill, a single family residence, and a decentralized network on the same platform. Further, we could self-custody these types of ownership claims in a single hardware wallet, if so desired. It means these assets to be able to reference each other contractually and interact in an automated way. It could mean global pooled liquidity for all asset classes through a single interface. Perhaps, it even means that we’ll hold less cash as working capital. Most of the major benefits of tokenization are dependent on the ability of computer systems and software to exchange and make use of information, or in a single word: interoperability. If we achieve full interoperability of assets, the ability to frictionlessly move value in and out of a diversified portfolio will have an impact on how we manage short term liquidity needs. Tushar Jain explores this idea here… Design space expansion recently reminded me of an anecdote about the history of television, summarized by Prof. Stevens of NYU: Lou Kerner here “Many early programs such as Amos ’n’ Andy (1951) or The Jack Benny Show (1950–65) were borrowed from early television’s older, more established Big Brother: network radio. Most of the formats of the new programs’ newscasts, situation comedies, variety shows, and dramas were borrowed from radio, too” Technology is often developed to improve something we already know and understand. What was probably less obvious at the dawn of television was that video content would evolve from a visual rendition of radio programming by a few major networks to millions of user generated videos on YouTube. The current stage of security tokens is analogous to broadcasting a radio program on television. We’ve just begun to tap the expanded design space for securities that it facilitates and we just don’t know how it will evolve from where we stand today. It could be a huge canvas for creativity over the next decade for those involved with security creation. Security tokens allow us to build in contractual features that have previously been infeasible to execute. It moves us closer to the economic concept of . complete contracts This is by no means an exhaustive list, but here are a few ideas to illustrate the point… >Ownership characteristics Features that reference the duration of ownership could be useful in shaping corporate governance and mitigating managerial myopia. An example currently being pitched by and others is tenured voting. The one sentence explanation is that the longer you hold the stock the more votes you get. This is tantamount to what founders are trying to achieve when they create a class of shares with 10x voting rights around the time they go public, but that’s a blunt instrument. It’s an interesting example of a contractual feature that would be easy to bake into a token. Cash flow rights (dividends) could also be altered according to ownership characteristics to create incentives for specific ownership structures. There’s a lot to unpack and think about in these models, but the point is that programmable securities will induce innovation in corporate governance. Eric Ries >Access rights The value of access rights doesn’t get enough attention. For example, many of Sequoia Capital’s venture funds have been notoriously difficult to access, even for some institutional investors. Sequoia has been very successful and there is evidence that performance within the VC asset class is persistent, perhaps . This suggests that access to Sequoia’s follow-on funds is a valuable asset. If you’re already an investor with Sequoia you will typically be offered an allocation of follow-on funds, but it’s not necessarily a contractual obligation, it’s based on the relationship. This means it’s not directly monetized. due to network effects If a tokenized ownership claim on a Sequoia fund contractually entitled the holder to participation rights in the follow-on fund, the token would presumably trade above net asset value (NAV) due to the value of the access rights. Let’s take it a step further and integrate cross-fund references. Perhaps a fund with lower demand, say Sequoia India, contractually includes a future allocation in the next flagship fund. Access that has historically been relationship based will move towards rights based once funds are tokenized, in order to unlock the value. There are many flavors of access rights. In addition to access to allocations of financial assets, security tokens could explicitly endow physical access to real estate, or admission to exclusive events. Minority owners of restaurants will gain access to priority seating or off-menu items. It could facilitate early access to research, or advance access to software releases. Access rights will integrate product markets with capital markets, merging the concept of “owner” and “user” not just for decentralized networks, but in the real world. Access to discounts could be conditioned on holding a specific set of assets, where the issuers have entered into a corporate partnership. In sum, access rights are valuable and will be monetized as they are built into security tokens. >Unbundling value Channeling Jim Barksdale, I would be remiss to offer examples of bundling, like attaching access rights to ownership, without acknowledging the design space around unbundling. Consider all the sources of value that could be unbundled within securities. Voting rights could be unbundled and sold to activists. Dividend rights could be unbundled like Treasury STRIPS. Companies could unbundle specific revenue streams and finance them independently. Complex revenue sharing and payment waterfall agreements become much simpler in this environment. There are lots of ideas to be explored in this category. >Cross-asset referencing Once asset classes are playing in the same sandbox we’ll see innovation in contracting across markets. I’m particularly intrigued by cross-referencing debt and equity during a liquidity squeeze. Essentially, I’m suggesting an automated restructuring feature. For example, say you have a mortgage on your house and have some equity value, but lose your job. Lack of income means you won’t be able to get a home equity loan to access that value and make your payments. In this scenario, you might default on your loan leading to foreclosure. If both the equity and debt were tokenized, a contractual feature could be designed where the mortgage token holder begins receiving equity tokens in lieu of payment as long as the value of the property exceeds the loan balance. There would no doubt be a penalty function since the mortgage holder doesn’t necessarily want equity, and would want to disincentivize the borrower from gaming the system. However, foreclosure is very costly for lenders and receiving value in the form of equity while keeping the owner-occupier in place as they try to resolve the liquidity squeeze may be the optimal solution. There are a lot of things to think through in this model, but it’s a fascinating concept. Mitigating foreclosure rates has the potential to unlock a lot of value, both financial and societal. There’s a lot of room for cross-collateralization beyond home equity. Loans will be secured by all sorts of things we haven’t considered, leading to credit expansion along many dimensions. Conclusion There’s plenty of work left to do, but the security token movement has substantial momentum at this point. The hype around blockchain will add fuel to this momentum, but only realization of real benefits will sustain it in the longer term. I’m organizing a in Manhattan on June 11 with , where I’m the Chief Strategy Advisor. We’ll explore the current state of the exchanges and trading platforms that are being built to address the considerable challenges that remain. If the challenges are met with solutions, there will be a strong case to create most securities as tokens someday. conference Security Token Academy At scale, we won’t call them security tokens, we’ll just call them securities. — - Further reading: (McKeon) Traditional Asset Tokenization (McKeon) Liquidity is about market depth, not magic (McKeon) The Promise of Automated Compliance (Pompliano) The Official Guide to Tokenized Securities (Koffman) Official Guide to the Security Token Ecosystem (Marks) The Future of U.S. Securities Will Be Tokenized This article benefited from discussing these ideas with many people over the past few months. I thank Ryan Alfred, Spencer Bogart, Andy Bromberg, Chris Burniske, Alex Evans, Matt Huang, Tushar Jain, Nikhil Kalghatgi, Lou Kerner, Michael Kogan, Dan McKeon, Ben O’Neill, Alex Pack, Anthony Pompliano, Bob Remeika, David Sacks, Derek Schloss, Darsh Singh, and Josh Stein for helpful comments. I will update this post from time to time to incorporate additional feedback. All errors are my own.