Blockchain ‘Promising But Inefficient’ in Payments

DNBBulletin, De Nederlandsche Bank | June 7, 2018

In a June 7th blog post, De Nederlandsche Bank (DNB) concluded that “for the time being” distributed ledger technology (DLT) is not suitable for its existing financial payment infrastructure due to its insufficient ability to scale for large volumes of transactions and other issues. DNB wrote:

DNB concludes that for the time being blockchain technology cannot meet the very high demands of a financial market infrastructure. This is evident from DNB’s own experiments with this technology. The biggest shortcomings are inadequate capacity, inefficiency due to high-energy consumption and lack of complete certainty about having paid a payment. Nonetheless, it appears that the resilience of a financial market infrastructure against external attacks could be increased by the use of blockchain technology, but this is at the expense of capacity and efficiency.

By way of experiment, DNB has developed and evaluated four prototypes with Distributed Ledger Technology (DLT) with the name Dukaton over the past three years. The aim was both to build up knowledge and to test to what extent these technologies are useful for improvements in payment and securities traffic.

A first prototype involved a Dukaton based on the Bitcoin software, with their own blockchain the Dukatons were created decentrally on five laptops via long-term calculations. Parties that validate the transactions to put in the blockchain receive a reward for creating new Dukatons and a transaction fee. In a second prototype, the Dukatons were created in a much less energy-consuming manner in advance. The creation of Dukatons, unlike with Bitcoin, was based on a trusted party that issues them. This party gets all Dukatons, while the parties that validate the transactions to put in the blockchain can only receive a transaction fee. Subsequently, a centrally set Dukaton was simulated in a third prototype, without Bitcoin software, but with a self-built electronic wallet (wallet). This stores the required cryptographic keys for security. This prototype then served as the basis of a fourth prototype to analyze the usefulness of DLT for a financial market infrastructure, such as a payment system. For this purpose, a website has been built that displays all Dukatons and transactions of the distributed ledger, and various configurations have also been tested. A number of algorithms have been used that approve transactions after a consensus has been reached between a number of nodes in a network. This prototype then served as the basis of a fourth prototype to analyze the usefulness of DLT for a financial market infrastructure, such as a payment system. For this purpose, a website has been built that displays all Dukatons and transactions of the distributed ledger, and various configurations have also been tested. A number of algorithms have been used that approve transactions after a consensus has been reached between a number of nodes in a network. This prototype then served as the basis of a fourth prototype to analyze the usefulness of DLT for a financial market infrastructure, such as a payment system. For this purpose, a website has been built that displays all Dukatons and transactions of the distributed ledger, and various configurations have also been tested. A number of algorithms have been used that approve transactions after a consensus has been reached between a number of nodes in a network.

The prototypes show that the blockchain solutions tested currently can not meet the high demands of financial market infrastructures (FMI). Requirements for FMIs are safety, reliability, efficiency, payment finality (legal security), authorization, resilience, availability, capacity, scalability, costs and sustainability. These requirements are high because FMIs play a central role in payment and securities transactions in order to carry out the settlement of transactions. An example of an FMI is Target2, the Eurosystem’s interbank payment system. The current payment systems are very efficient, can handle large volumes and provide the legal certainty of having paid. The blockchain solutions tested show that they are not sufficiently efficient, with regard to costs and energy consumption, and they can not handle the large numbers of transactions. Moreover, with some used consensus algorithms the 100% certainty is never achieved that a transaction can not be reversed, whereas central banks do offer that with Target2. There are algorithms that are resistant to malicious parties and can increase the cyber resilience of FMIs, but these algorithms currently do not meet the other requirements imposed on FMIs. DLT may offer improvements in the efficiency of payments involving different currencies.

DNB finds the technology behind Bitcoin, the blockchain, interesting and promising. Perhaps new algorithms will meet all the requirements for FMIs in the future. That is why DNB continues to invest in knowledge of this technology and is experimenting with it. DNB also conducts discussions with new and existing parties in the market about possible applications, in order to contribute to innovation from its role.