Executive Summary

Building Blockchains stat bar

Earlier this year, Greenwich Associates released research looking at the current state of enterprise blockchain adoption in capital markets1. The study found distributed ledger technology (DLT) solutions were steadily moving along the adoption curve, with 14% of companies reporting they had already moved into production. It is fair to say, though, that the industry has lagged behind its own optimistic expectations from two years ago. Implementing enterprise technology designed to replace decades of legacy market infrastructure is no simple task, and 57% of blockchain executives told us it has been harder than expected.

In this latest research paper, we look at some of the key technical challenges executives are facing and analyze their approach to solving them.

Building Blockchains stat bar

Earlier this year, Greenwich Associates released research looking at the current state of enterprise blockchain adoption in capital markets1. The study found distributed ledger technology (DLT) solutions were steadily moving along the adoption curve, with 14% of companies reporting they had already moved into production. It is fair to say, though, that the industry has lagged behind its own optimistic expectations from two years ago. Implementing enterprise technology designed to replace decades of legacy market infrastructure is no simple task, and 57% of blockchain executives told us it has been harder than expected.

In this latest research paper, we look at some of the key technical challenges executives are facing and analyze their approach to solving them.

Scalability

One of the most discussed problems with DLT is the relatively low transaction processing speeds. For example, the bitcoin blockchain can only process about 7 transactions per second (TPS) compared with tens of thousands TPS for major credit card networks. However, the bitcoin blockchain is slow primarily because the network uses a proof-of-work consensus algorithm. This is the blockchain’s secret sauce: Validating nodes are required to solve complex and time-consuming mathematical puzzles in order to protect the integrity of the public network.

In contrast, enterprise blockchain solutions are trusted private networks and can instead use a different, faster kind of consensus algorithm. (See the Greenwich Report Securing the Blockchain2 for a discussion on this topic.) But even in an enterprise solution, the ledger is distributed among a broad network of geographically diverse participants who must communicate with each other when validating transactions. This is inherently slower than a centralized system.

We find scalability to be one of the top concerns among firms implementing DLT solutions, with 42% of respondents overall citing it as a major issue.

Top Issues in Implementing DLT Solutions - Scalability

It is striking to note, however, that blockchain technology companies feel scalability is not a major issue at all. This is in sharp contrast to the other firms in our study, which are, presumably, the clients of the blockchain technology companies. Over-optimism on behalf of these technology providers could explain this disconnect. Another plausible reason could be that real-world implementation of DLT solutions significantly impairs performance compared to lab-run simulations.

DLT Transactions Per Second

Drilling into the scalability issue, we asked respondents to provide specific data on the transaction performance speed they were experiencing. We note that most firms are still in the development stages and have not moved to production. Over one-fifth have not been able to surpass 100 transactions per second in this environment. At the other end of the spectrum, it is encouraging to see that 2% of firms have moved into production and are achieving high levels of throughput. While this may seem like a small number, it is a positive sign that competitive transaction speeds are achievable with DLT solutions.

The Payments Leg

Just as there are two parties to a transaction, there are also two “legs.” Typically, one party agrees to exchange an asset in return for payment from the other party. The asset can be equities, bonds, commodities, derivatives, etc., and the payment is typically in cash (fiat currency). So every blockchain project looking to build a new market structure for any asset will need to figure out how to handle the payments leg.

Synchronizing the asset trade and the payment is important to ensure settlement finality—delivery versus payment. While many DLT solutions seek to achieve settlement in t+0 (same day), this cannot be possible if FedWire takes 24 hours or more to transfer U.S. dollars, and Automated Clearing House (ACH) takes two to three business days.

The most popular solution to the payments leg is to use standard financial rails, such as FedWire, ACH or bank APIs. This is often the easiest solution as well, as it requires minimal additional development or collaboration. A good example is the Nasdaq Linq platform. Nasdaq uses the Chain blockchain ledger for private securities transactions and has integrated with Citi’s CitiConnect® payments API.

Handling of Payment Leg

As noted, using standard financial rails may mean settlement has to be slowed down for some DLT applications but can still represent a significant improvement in many use cases. In addition, it makes sense for firms to focus on implementing DLT for the financial asset first, while using traditional methods for payments. At a later stage, companies can decide if they want to use a different solution for the payments leg.

For companies seeking an alternative to traditional financial rails, there are a few different options being pursued. Some firms are working on a DLT solution for the payments leg. This will enable them to make payments with asset settlement.

So-called “settlement coins” are an innovation designed to tackle this exact problem. Goldman Sachs was granted a patent on its SETLcoin last year. The biggest initiative in this space is the Utility Settlement Coin (USC) project, which was launched by UBS and blockchain technology company Clearmatics. It has now grown into a project involving 17 different consortium members.

A settlement coin is a digital asset that can be used to represent fiat currency in financial transactions between members of the consortium. The idea has been around for a while and has the backing of many leading banks, so it is somewhat surprising that only 15% of firms are looking to use this method for the payments leg.

Almost as popular for payments settlement are “IOU tokens.” Like settlement coins, these are also digital assets but are issued by the blockchain technology company. They reconcile separately with traditional financial networks. Because the coins can only be issued by the blockchain technology company and do not require collateralization or industry backing, they are quicker to implement. However, they do not represent a long-term remedy to the problem and will likely be phased out as other more robust solutions become available.

Arguably, the most potent solution would be a central-bank-backed digital currency—CBDC. Think of this as Fedwire on a blockchain or “Fedcoin.” CBDC do not exist yet, although many central banks are working on initiatives in this space3. Fourteen percent of companies told us they are working with central banks on this type of product.

A small minority, 4%, are looking to use public cryptocurrencies such as bitcoin or ether to represent the payments leg. This is unlikely to be a viable solution due to the volatility of these assets—and because they are still not accepted within the financial system.

While public cryptocurrencies are not a likely option for handling payments, there is a new kind of solution, based on public blockchain networks, that has been announced in recent weeks. Two firms, Gemini and Paxos, both hold banking charters with the New York Department of Financial Services and have received approval to issue stablecoins. These tokens will have a fixed price of $1 (hence the name stablecoin), and each token will be backed by actual dollars held at trust banks. The tokens are also ERC 20 tokens, meaning they run on the Ethereum blockchain network. It’s too early to tell if these new stablecoins will gain traction in financial services—but as a product with rapid settlement, U.S. dollar fungibility and backing by a regulated trust bank, it checks many of the boxes.

Transaction Confidentiality

It is well known that public blockchains like bitcoin contain a complete record of every transaction, which can be queried and viewed by anyone with a browser. While this transparency may be appropriate for cryptocurrencies, it is a less desirable feature for an enterprise blockchain. In many markets, participants do not want the details of their trades immediately visible to all other participants. For example, Goldman Sachs may not want Morgan Stanley to know that it just executed a big trade in an illiquid security at a certain price—particularly if there are likely to be follow-on transactions.

For this reason, almost two-thirds of executives believe that incorporating zero-knowledge (ZK) proofs, or similar technology, are an important part of an enterprise DLT solution.

Importance of Zero-Knowledge Proofs

ZK proofs are a recent innovation. They require an additional layer of cryptography in the consensus protocol that allows one party to prove to another that something is true without revealing any other information.

Hardware Security

As with any financial network, security is always going to be a major concern. There have been numerous instances of hackers stealing private keys and looting billions in cryptocurrency, and other widely publicized hacks of traditional financial networks.

As part of the security framework for enterprise blockchain solutions, almost half the firms we spoke with are or will be implementing hardware security modules (HSM).

Using Hardware Security Modules

An HSM is a physical computing device that secures and manages digital keys and provides strong authentication and encryption. HSMs allow for the generation, storage and management of secure cryptographic keys that are used to sign blockchain transactions. They are proven solutions that are currently used in retail banking and in payment transactions.

Blockchain Edits

A key feature of blockchains is that the ledger contains an immutable, censorship-resistant record of every transaction. Each transaction block is linked to the previous transaction block—all the way back to the very first. This immutability is generally seen as a strength, as it provides built-in audit functionality and tamper-resistance.

However, immutability also means that corrections can’t be made in the event of errors or fraud. As a solution, Accenture proposed (and was awarded a patent for) editable blockchains. However, only 11% of firms are exploring this type of technology.

Exploring Editable Blockchains

Many people working on DLTs had previous experience with bitcoin or other cryptocurrencies. For them, the idea of a mutable blockchain may seem like a contradiction in terms. However, the problems that an editable blockchain seeks to solve do exist, and firms will need to figure out alternative solutions. In the case of a trade error, for example, rather than editing the ledger, a separate transaction would need to be executed to reverse the erroneous trade, followed by a new transaction representing the correct one.

It has also recently emerged that the immutable nature of blockchain may be in contravention of the EU’s General Data Protection Regulation (GDPR)4. Article 17 of these rules provides for “the right to be forgotten.” The issue has arisen from the desires of individuals to progress in life without being permanently characterized by actions from the past. Right-to-be-forgotten policies were conceived in response to the raft of information contained on the World Wide Web, but the actual regulations are much broader and require companies to be able to erase personal data upon demand.

This is where the conflict with an immutable ledger can arise. One solution to this problem is to assure that personally identifiable information does not get onto the blockchain in the first place. As the GDPR regulations only came into force in May 2018, it is still unclear exactly how the industry will adapt.

Final Thoughts

Within financial services, there is growing concern that blockchain will not be able to deliver on the promise of revolutionizing markets. We have entered a classic trough of disillusionment—a phase of the Gartner Hype Cycle that characterizes the frustration and waning interest that occurs as projects take longer to get off the ground and fail to deliver on the hyped expectations.

Indeed, a majority of executives working on DLT initiatives have told us that implementation has been harder than expected. However, our research also indicates that some teams are resolving key issues such as scalability, and addressing others—such as transaction confidentiality and security—using a combination of new techniques and tried-and-tested technologies.

Richard Johnson


¹https://www.greenwich.com/equities/blockchain-adoption-capital-markets-2018
2https://www.greenwich.com/fixed-income-fx-cmds/securing-blockchain
3https://www.greenwich.com/blog/digital-currency-banks-can-get-behind
4https://eugdpr.org/

Methodology

During February and March 2018, Greenwich Associates interviewed 213 global market participants working on blockchain technology to assess the current state of blockchain adoption in capital markets. Respondents include representatives from a broad array of organization types, 93% of whom are either key decision-makers or actively involved in blockchain initiatives.

Organization Type and Respondent Role