Security Tokens (Part 2)

12 min readJan 14, 2019

In Part 1 of this post we have discussed the different ways of how investors can hold securities, notably in paper-based bearer format, in centrally-registered dematerialised format (the current default), and in tokenised aka blockchain-registered dematerialised format. We have also identified and defined a number of dimensions in which those formats differ. Those dimensions were

Custody
Registration
Liquidity
Ease of settlement
Payments
Communication
Authentication
Fungibility
Traceability
Tainting
(Im)mutability
Recovery
API access

In this part of the blog post we discuss the points 1–7. The remaining points will be discussed in the forthcoming Part 3.

Note: Colin Platt commented on a pre-view version of this post; I have decided to include his comments as notes and address them explicitly as I think they make very interesting discussion points

Custody and Registration

Custody. For physical securities, custody refers to where the actual pieces of paper are held, how their owner is registered, and how they can be transfered to a new owner. For dematerialised securities custody only refers to the second part of that.
Registration. Registration refers to the registration of the investors, allowing the company in particular to contact them and to understand their identities.

Custody and registration are the dimensions in which the three different forms of holding a security differ most. Let’s start with paper-based securities: they allow for self-custody, ie the investor can simply hold the securities in their safe at home. In this case a priori there is no share register, and the company makes announcements of corporate events through well-known channels, typically the financial press. It is then the investors’ responsibility to take action. For example for dividend payments the certificate contains coupons (one per payment) that the investor has to clip and submit for payment.

Paper-based securities also allow for third-party custody where the custodian is simply a service provider who keeps the securities in their safe and who takes care of all the administrative tasks like clipping and submitting the coupons for payment, informing the investors of AGMs etc. This makes the investors’ lives easier, but it comes with an important drawback: the investors must trust their custodians because other than the legal system in which they are embedded there is no guarantee that the custodians not run away with the assets.

Today most securities are held in dematerialised format, which is a fancy way of saying that the information of who owns how many shared is held in a big spreadsheet, or more realistically in an SQL database that is run by a designated entity. As far as customers are concerned this is exactly the same as custody of paper securities, except that processes are typically automated and faster, and typically also available via a web interface and/or an API.

Crypto assets allow for a much wider spectrum of custody solutions. First of all there is again self-custody like with paper-based assets where the investors holds the private keys themselves, allowing them to run all the operations from their own devices. Alternatively there is a standard custody solutions where the custodian holds the private keys, and which from a user perspective is hard to distinguish from a custodial solution in the traditional world.

There is one important difference though: when holding dematerialsed assets in custody there is one blessed database that contains the authoritative information about the share ownership, and every custodian must be connected to this blessed database. This often poses a problem for cross border transactions, especially for smaller investments and secondary countries: whilst a billionaire will always find a way to custody an asset they want to invest in via their Geneva-based private bank, the German dentist who wants to invest €10,000 into shares listed on the Uruguay growth market might find a hard time finding a custodian who is technically able to do that, and even less so at a reasonable price.

Note: Colin points out that that this might simply be a question of demand and unit cost economics — if people wanted those pipes they’d probably be built. I agree, but my point is that unit cost economics and friction more generally matter. Essentially it is the “the iPhone is just a Nokia with a worse camera” argument — yes, my hypothesis here is that because blockchain pipes are cheaper / have less friction they might enable business that does not currently makes sense.

Crypto custody on the other hand is key custody: by design the blessed database (aka blockchain) is world-writeable provided one has the correct private keys, which means our German dentist can simply use his regular German (crypto-)custodian for all his crypto assets, be they located in Germany, Uruguay, or North Korea.

Crypto custody however can even go further. For example it can be designed in a manner that both the investor and the custodian have a key, and that the investor keys allows for communication and voting, whilst for transfering the securities both the investor and the custodian keys are needed. This setup provdes additional security. This however is a complex topic that merits its own blog post, especially when we want to take the risk of losing one’s private keys into account.

Ease of Settlement

Ease of Settlement. Ease of settlement refers to how hard (or easy) it is for a buyer and a seller to complete the ownership transfer.

One of the key advantages of security tokens over all classic alternatives is the ease of settlement. Not only is extremely easy (signing the transfer with the correct private key), but also settlement is nearly instantaneous, and there is no issue with crossborder settlement. Also very importantly it is relatively easy to avoid settlement failures, at least catastrophic ones of the kind where only one leg of the transaction goes through (see the section on clearing below). Transactions can of course still fail to settle, but this is relatively painless in an environment with a very short settlement cycle where the risk of massive market moves is not that big.

Note: Colin points out that “there is no issue with crossborder settlement” is not quite right as for example it ignores all the potential legal issues that come with this. Fully agreed, my point was solely about the technical ability to settle cross-border trades very cost effectively. The legal issue needs to be looked at in detail, and incidentally one of my core beliefs here is that the technology part of securities tokens is the easy part and mostly solved — it is the legal part where we need to focus our R&D effort as in industry.

For dematerialised securities, settlement can be equally easy if both counterparties have accounts with the central registry. Especially smaller holders (or those who want their positions to be secret, like hedge funds) might prefer to have a custodian hold those assets on their behalf. In this case, the two counterparties must work out who their respective custodians are, transfer the securities over that the custodian level (possibly netted with the flow of other customers), and finally the custodians must adjust their internal records accordingly. Each step can go wrong — and statistically does go wrong given the numbers involved — so there will be a number of reconciliation and exception handling procedures involved depending on where exactly the process fell over. Also cross-border settlement is hard, and investors might need a different custodian for the different countries where they invest because not every custodian can hand all securities.

For physical securities finally, settlement literally involves handing over the actual pieces of paper, either from investor to investor in case of self-custody, or from the custodian of one investor to the custodian of the other investor. Again, many things can go wrong in this process, and assuming the consideration is paid via bank transfer, there is always a risk that one leg of the deal fails whilst the other one settles. Cross border settlement (or even long-distance settlement within a country) is often prohibitively expensive, and subject to risk of less and theft.

Clearing

Clearing. Clearing refers to the process that ensures that trades always settle correctly, typically through a trusted third party.

In the classic securities clearing process that is now standard for most regulated exchanges throughout the world trades are novated, meaning a third party — referred to as the clearing house or central counterparty (“CCP”) — inserts itself into the deal, becoming the buyer to every seller and the seller to every buyer. This ensures that both counterparties can rely on that trade going through as negotiated, regardless of counterparty performance. This is important for the liquidity of the market as participants do not have to assert the credit-worthiness of their counterparty before trading, which in turn is an issue with automatic matching.

The way it works is that the CCP takes delivery of the money and the securities respectively and passes both on at the agreed date, typically T+2 ie two days later. If the buyer fails to pay the CCP simply takes delivery of the security, pays the agreed price for it, and sells it in the market. Similarly if the seller fail to provide the security the CCP purchases it on the market and sells it to the buyer. Any loss made is taken out of the defaulted counterparty’s margin account

Note: a margin account contains the money that every market participant in a cleared system must post before doing any trading and that covers potential losses from scenarios like the one above. For market participants accessing the market through a broker the latter is required to post a margin, and it is a commercial decision whether or not to pass that requirement through to their often retail customers or to deal with loss recovery ex post

Clearing prevents two risks: on the more fundamental level, it prevents theHerstatt risk of handing over securities without receiving payment or vice versa. On a higher level — and that only works in sufficiently liquid markets — it ensures that the transaction goes through as agreed because the CCP just sources the security from the market, or sells it into the market.

For tokenised securities smart contracts can easily fulfill the first function: a fully audited contract can, after being notified of the terms of the trade by either the exchange or the two parties, take delivery of the cash and security tokens and forward them as required once both legs are complete; if one party fails to post the agreed assets the other one can simply withdraw theirs. The second function is harder to perform, especially in today’s crypto markets that are often too illiquid to reliable source or sell asset on an automated basis.

Payments

Payments. Payments refers to how the payments in respect to the security are handled.

For standard dematerialised securities, payments have to go through the traditional banking system. What this means in practice is that there is a payment agent who is in charge of coordinating the payments, and that someone along the custody chain must know the customer’s account details, typically the custodian closest to the customer who depending on business model and jurisdiction might or might no be running a cash account for them as well. Payments are then cleared along the traditional clearing pathways, with all the costs and problems this involves. On top of this there is always the risk of having wrong bank details on file, eg because someone changed accounts or died, and having to deal with those is costly.

For paper-based securities the process is even more cumbersome and archaic: Those securities have coupons for every payment, so the customer (or their custodian) has to literally clip the coupon (which is where that expression comes from) and submit it to the paying agent. They then might receive a cheque that is cleared through the banking system, or — in a slightly more modern scenario — provide their bank account details and receive the money via bank transfer.

When using tokenised securities on the other hand payments can be made directly through the token. Until recently it was only possible to make those payments in crypto currency — typically the native crypto currency of the network, for example Ether — but since the emergence of regulated stable coins it is possible to make quasi-USD (and soon other currency payments) on-chain which are automatically delivered to the holder of the token, meaning all the backoffice work around determining bank details and exception handling is no longer necessary.

I can’t be stressed enough how important the emergence of reputable stable coins is in this context: most people in the real world do not want to interact with crypto assets like Ether and Bitcoin. In principle payments could be made in crypto assets and both the payer and the payee take care of the fx from and into fiat if they so wish, but this has a number of drawbacks. Most importantly banks are not currently to keen on interacting with people who deal in significant amounts of crypto currencies, which poses both a problem for the companies who have to convert fiat to crypto and the investors who have to convert it back, especially if the amounts are substantial. Also there is the volatility risk in the period between when the company exchanges fiat into crypto, and when the investors exchange it back. With bad timing the volatility in this period can be substantial, especially if the process on the investor side is not automated.

Communication and Authentication

Communication. Communication refers to authenticated communication between the issuer of securities and the investors. This category also includes voting.
Authentication. Authentication refers to how the owner of an investment contract can prove their identity.

The communication issue is similar to the payment issue: to the extent that personal communication from the company to the investors is required, in the current environment someone must hold the investors’ contact information on file and then send out all those letters and/or emails. Reverse communication (from the investor to the issuer) is even harder because it needs to be authenticated. Typically this means there needs to be a signature on file against which the communication must be validated, which at the same time is cumbersome to execute and does not provide a lot of security, especially in combination of electronic transmission of documents via fax or scanned pdf.

Again in the blockchain world the communication can go directly through the token. In its easiest form it is a public broadcast, simply notifying investors that a letter is available, which is sufficient for things like invitations to an AGM. For private communication the transmission can be encrypted without the need for an additional key infrastructure because every crypto address corresponds to a private key that can also be used for encryption. This key can also be used for signature and authentication which makes reverse communication much more secure than a regular signature would. This feeature makes it relatively straight forward to implement features like voting in a virtual AGM.

Liquidity

Liquidity. Liquidity refers the ease with which someone who wants to sell can find a buyer and vice versa. Liquidity depends on the size of the position the buyer or seller wants to shift.
Ease of Settlement. Ease of settlement refers to how hard (or easy) it is for a buyer and a seller to complete the ownership transfer.

Liquidity as defined above is a priori not dependent on the format in which a security is held, it depends first and foremost on the asset itself. For example a share of global company like Apple will be massively more liquid than that of a small company in a small country, and a company’s equity is more liquid than any of the many bonds that company has also issued.

Secondly liquidity depends on the market: companies that are listed on one of the major markets like NYSE, LSE, or Deutsche Börse will be more liquid than those listed on secondary exchanges who will in turn be more liquid than those not listed at all. Now format does play a role here, but only a secondary one: currently it is a listing requirement for regulated exchanges that the securities are held in dematerialised format with a centralised depositary. Arguably up to now this made sense: settlement of paper-based securities is complicated, lengthens the settlement cycle, and error can undermine the confidence in the market. However, capacity issues on chains like Ethereum aside, there is no intrinsic reason why regulated exchanges should not settle via a blockchain, and therefore in this case the format matters for liquidity, but only for legacy reasons.

An important dimension we discussed above is the ease of settlement and this can be substantively better for tokenised securities than for any other format, in particular in a cross border context. In fact, as discussed under the custody point, investors other than the super-rich can often only hold assets in a very limited number of jurisdictions, so ease of settlement is in many cases zero: it is simply impossible to transfer that security to an investor living in the wrong jurisdiction. Removing this constraint can improve the liquidity of an asset: if there are suddenly many more people who can in principle buy an asset there is a decent chance that there are also some of those who want to buy that asset.

In this sense, tokenisation can increase the liquidity of an asset. However, it is a secondary effect, and it still requires additional infrastructure, notable something akin to a listing, for buyers and sellers to find each other.

this is the end of Part 2; continued with Part 3

Stefan Loesch a managing partner at LexByte, an advisory firm specialising on tokenised investments. He has more than 20 years experience in financial markets, and his previous roles include advisory at J.P. Morgan and McKinsey and quant development at Paribas. He is the author of “A Guide to Financial Regulation for Fintech Executives” (Wiley 2019).