The Finternet: A giant leap for the financial system

Speech by Mr Agustín Carstens, General Manager of the BIS, at the Peterson Institute for International Economics event on "A vision of the future financial system", Washington DC, 19 April 2024.

Good afternoon.

I would like to thank the Peterson Institute and my good friend Adam Posen for hosting us today. I am delighted to be joined by Nandan Nilekani, my collaborator in this endeavour. And I look forward as well to hearing from Roberto Campos Neto, Joachim Nagel and Changyong Rhee – three central bank Governors whose institutions are pushing the boundaries in the use of technology to drive financial innovation.

Today, Nandan and I are proud to present our new report, "Finternet: the financial system for the future". This is the result of more than a year's worth of discussions and thinking since we first met to brainstorm the concept on the sidelines of the G20 meeting in Bengaluru in February last year.

For the BIS, this is another stage in our exploration of how advanced technology could deliver a more innovative and efficient financial system. I am encouraged by our work in this area. We have learned many valuable lessons and drawn some striking analytical insights.

But at the end of the day, we at the BIS are mostly economists and finance specialists. To really understand how to apply technology, we need to bring technologists fully into the conversation.

That is what Nandan and I have sought to do. We aimed to bring together the economic and financial perspectives we have at the BIS with the technological know-how of Nandan and his team. And, in doing so, to articulate a vision for the future financial system that unites a robust economic and financial foundation with state-of-the-art technology. In particular, we sought to sketch a blueprint of the economic and technological architecture for a future financial system and show how we could turn it into reality.

Nandan and I come from different worlds. I wouldn't go so far as to say that economists and technologists speak different languages, but we have our own distinct dialects. At times there was dissonance in views and perspectives. This is healthy. If we don't start the dialogue between economists and technologists, it is hard to see how useful financial innovation can really move forward.

We have labelled our vision for the future financial system the "Finternet". I will leave it to Nandan to expand on its more technical aspects. But let me provide a broad overview.

The Finternet is a vision of multiple financial ecosystems that connect with each other, much like the internet. It aims to empower individuals and businesses by helping them to have full control of their financial lives. The application of technology has meant that tasks that were once expensive and time-consuming, like making an overseas phone call or booking a hotel room in an unfamiliar city, can now be done with the flick of a finger. The Finternet aims to bring the same kind of progress to financial services. We foresee a system in which individuals and businesses could transfer any financial asset, in any amount, at any time, using any device, to anyone else, anywhere in the world. Financial transactions would be cheap, secure and near instantaneous. And they would be available to anyone.

This is particularly important for emerging and developing economies. Despite many noteworthy initiatives to bolster financial inclusion – spearheaded by sovereign authorities, central banks, multilateral financial organisations and the United Nations – large gaps in access to financial services between advanced economies on the one hand, and emerging market economies and least developed countries on the other, still exist. It is striking how many financial services and instruments are just not available in many of these countries, particularly to people living in remote areas, or who have low incomes. And even when residents can access financial services, high transaction costs and slow services are pervasive. These reflect a number of structural deficiencies, including a lack of competition, inadequate public infrastructure and bottlenecks induced by legacy systems that have not kept pace with the capacity needs of a modern economy.

Our starting point was the observation that recent years have seen a flowering of innovation in financial technology. This has led to some important breakthroughs.

Programmable ledgers are one example. These digital platforms combine the record-keeping functions of traditional databases, used to track the ownership of financial assets, with the rules and governance arrangements needed to update the databases. The ledgers enable "smart contracts". These are self-executing applications that can trigger an action, for example the sale of an asset, if a pre-specified condition is met. They have the ability to bundle many of these automated transactions together, which is known as "composability".

Tokenisation is another example. Tokens are the digital assets that exist on programmable ledgers. They contain both the information necessary to uniquely identify the assets and their owners, and the rules and logic governing their use. 

And more advances could soon arrive. I am particularly encouraged by the transformative potential of artificial intelligence for the financial sector. The BIS will have more to say on this when we release our Annual Economic Report in June.

As we reflected on these developments, Nandan and I were both struck by the focus of innovation efforts on very specific technologies, financial assets and use cases. Surprisingly little thought has been given to how technological innovations could fit together and contribute to public policy objectives.

This matters because the real value of many of these technologies and applications will only come when they are combined into a seamless, interconnected network. To take one example: tokenised assets could drastically lower the reliance on lengthy messaging, clearing and settlement processes that are the cause of so many costs and delays in today's financial system. But they will only do so if different asset tokens – including tokenised forms of money – exist on the same digital ledger, or at least ledgers that can seamlessly interact with one another. Similar "network effects" apply to many of the other financial innovations.

As an analogy, I think you will all agree that mobile phones are much more useful than they were a decade or two ago. This is because of advances in all aspects of phone technology: in the handsets, in the quality of apps and in the shift from 3G to 5G networks. The combination of these advances is worth much more than the sum of their parts. A 5G network would be of little use if we still had the same phones as at the turn of the century. And smartphones would be wasted if we could only use them to make phone calls and send SMS messages.

The principle behind the Finternet is similar. To really unlock the value of financial innovation, we need to bring all of the pieces together and seek to dismantle barriers and silos that exist in the current financial system.

And it is important to do so now, when the technology is mature enough to use, but before we get locked into rigid institutional frameworks or walled gardens. We have a once in a lifetime opportunity to revisit the architecture of the financial system. Instead of asking what financial technologies we will use in the future, we should ask what the entire future financial system will look like. And we should figure out how we can make a quantum leap to get there.

So what do Nandan and I propose?

The application of more advanced technology is necessary. But technology is not an end in itself. Its value comes when it is applied alongside robust economic and financial architectures and trusted institutions.

The core idea behind our approach is to have a financial system that relies on the exchange of tokens in the context of unified legers. Intermediaries would facilitate the trades, the onboarding, the offboarding and the tokenisation of assets. The management of tokens would be done by the different intermediaries and ledger owners that will be a part of the system. Key procedures would be the authentication of the parties (who is trading), the verification of ownership (who has what), and the off- and on- ramps, as they are key to integrating the system with the rest of the economy. In short, tokens and unified ledgers based on programmable platforms enable a shift in how financial transactions would be conducted.

This will require a number of building blocks.

We see unified ledgers – programmable platforms that bring together all of the financial assets needed to complete a financial transaction – as an important vehicle to deliver the Finternet. Equipped with tokenised assets, these ledgers would drastically reduce the need for lengthy messaging, clearing and settlement systems. They would enable programmable transactions and greatly simplify compliance processes. They would meet – and perhaps exceed – the regulatory and supervisory standards that are required in today's financial system. In doing so, they would unlock the power of tokenisation, offering users faster speed, lower costs and more reliable service.

To be clear, we are not advocating for "one ledger to rule them all" – a single ledger that encompasses all financial assets and transactions in an economy. Depending on the needs of each jurisdiction, multiple ledgers could co-exist. They would need to connect with each other, and with other parts of the financial system, for example through application programming interfaces. And their function, and the range of assets they contain, would probably evolve over time. They might start with a small number of assets and use cases before gradually expanding into something more ambitious.

A two-tier monetary system – similar to the one we have today – will be another important building block. Central bank money will represent the first tier, and commercial bank money the second.

Settlement on the central bank balance sheet is the ultimate guarantee of finality in financial transactions. As such, wholesale tokenised central bank money is a necessary foundation piece. It will play a similar role to reserves in today's financial system, but offer enhanced functionality. Some central banks may also consider issuing retail tokenised central bank money – a digital equivalent to banknotes – to provide even more choice to consumers. I would emphasise, however, that traditional payment methods, including notes and coins, would remain available for those who wanted to use them.

Tokenised deposits will provide the natural retail counterpart to tokenised central bank money. As in today's financial system, commercial bank money will serve as the primary means of payment for most individuals and businesses.

In passing, let me mention that earlier this month we at the BIS launched Project Agorá in collaboration with seven leading central banks and the Institute of International Finance. This project will investigate how tokenised commercial bank deposits can be seamlessly integrated with tokenised wholesale central bank money in a public-private programmable financial market infrastructure, with a specific application to cross-border payments.

A third building block for the Finternet will be a robust regulatory and supervisory structure. It will preserve the institutional arrangements, including deposit insurance, settlement on central bank balance sheets and prudential regulation, that today ensure the singleness of money and trust in the financial system more broadly. And it will ensure that the Finternet does not provide a venue to circumvent laws or to engage in regulatory arbitrage. Meanwhile, the use of technology in supervision and regulation could lower compliance costs, while delivering stricter adherence to rules and regulations.

In the Finternet, current financial services will still exist, but be more efficient. And the combination of robust economics, innovative technology and strong governance will create space for new, innovative financial products to emerge. In economics jargon, markets will be more complete.

Meanwhile, by lowering costs and creating new venues for financial activities, the Finternet will lower the barriers that still keep far too many people locked out of formal financial services. The gains could be particularly large for emerging market and developing economies.

So, how can we make the Finternet a reality?

A first step is to clear the impediments that currently stand in the way. Laws and regulations that restrict the deployment of new technologies are one example. The ambiguities that still exist in many jurisdictions about the treatment of tokenised assets and other innovative financial products deserve urgent resolution.

A second step is for both the public and the private sector to play a part. Public authorities have an important role to play. Through the development of digital public infrastructure, they can establish the platforms, rulebooks and regulatory protections required to deliver an open and efficient financial system. And as suppliers of central bank money, they will continue to provide the foundational asset for the entire monetary and financial system. But public authorities cannot and should not be the main providers of digitally advanced financial services. That is the private sector's job. They will both need to collaborate to realise our vision of the Finternet.

Finally, there is an urgent need to experiment to learn more about how innovative technologies like tokenised assets and unified ledgers can be deployed in the real world. At the BIS, we are doing our part, not least through our Innovation Hub. There, in conjunction with our partner central banks, we are pushing the envelope and exploring what is possible with innovative financial technologies.

There is plenty of space for others to join us. I truly believe that now is the time for a "Neil Armstrong moment" – that small first step that represents a giant leap for the financial system. We know where we need to go. We have the tools to get there. Now is the time to take the first steps. Let me now hand over the floor to Nandan, who will tell us how this vision could be achieved in real life.