Below is a summary and select slides of Adam Ludwin’s keynote address to a gathering of ~100 central bankers and regulators from around the world at the Federal Reserve in Washington D.C., jointly hosted by the IMF and World Bank. The event, titled “Finance in Flux: The Technological Transformation of the Financial Sector,” was held on June 1st, 2016. This presentation was preceded by an inaugural address by Federal Reserve Chair Janet Yellen.
I’d like to start by thanking Chair Yellen, the staff of the Federal Reserve, and the conference organizers for the invitation. It is a pleasure to be here today with such an incredible and international group.
Today’s talk is titled “A New Medium for Money” because that is what a blockchain enables. Chain partners with organizations like Visa, Citi, Nasdaq, Fidelity — and increasingly with central banks — to deploy next-generation financial networks on blockchain architectures.
To put this talk in perspective, it is useful to think back to November 2008. It was just two months after Lehman Brothers collapsed. AIG and other financial institutions received $200 billion in federal assistance. The Fed announced an $800 billion stimulus program and Obama won the presidency. It was a busy month in the news.
But something else happened that month that didn’t make the news. In an obscure corner of the internet, an anonymous person or persons published a math paper — the “Bitcoin white paper” — that solved a problem that had until then stumped computer scientists: how to create digital money without any trusted parties. The key breakthrough was the invention of a cryptographic data model called a blockchain.
Recall that, at the time, the financial system was experiencing a run on the banking system. The credit crunch was made worse by a lack of clarity around asset ownership. The crisis, and its causes, were not lost on the author of this paper. In fact, he or she cited the financial crisis directly by embedding a link to a Times of London article about bank bailouts in the first Bitcoin transaction.
Seven years on, the Bitcoin network has proven to be robust, with the volume of Bitcoin transactions now over 250,000 per day. While few noticed Bitcoin before 2012, it is now evident that the model is resilient, despite all the controversy.
Bitcoin created digital money by creating an entirely new asset class. There are no network operators and no central issuing authority. Bitcoins are issued by the network at a predetermined rate that will max out at 21 million total units.
So how does Bitcoin relate to the efforts now underway by financial firms and some central banks to create digital money? After all, they are not using Bitcoin directly, so what are they doing?
Simply put, these institutions are using the techniques pioneered in Bitcoin to build new networks that digitize existing asset classes, like securities and currencies, so they can move more efficiently and securely. The reason institutions are building new networks, instead of just using the Bitcoin network, is because the Bitcoin network is not designed to support the issuance of assets like stocks, bonds, and currencies from a technical or governance perspective. It is designed to issue and transfer bitcoins.
Post-2008, regulators put rules in place to improve the safety and soundness of the financial system. But an unplanned innovation was also born out of the crisis. It began at the periphery but is now moving towards the core of how financial markets work.
Blockchain: a new medium for money
To summarize the big idea of a blockchain: it enables a new medium for money. One that is in a native digital format.
New mediums almost always change a market’s structure. Consider how changing mediums have transformed other industries, like music, publishing, and communication. As the medium of music went from live, to recorded, to digital and now to streaming, the market structure of the music industry changed quite dramatically, with new winners and losers at each stage. The same is true in publishing: from written text, to bookmaking, to the web, and now mobile/tablet. We find new players and distribution models in each stage. Finally, in communication, we have gone from being limited to talking with those near us, to the telephone system, to voice-over-IP. Long distance phone rates and the business models that depended on them no longer make sense when our voice travels in a digital medium over the internet.
So what does this have to do with money? Well, the medium of money has only changed a few times in history, from precious metals to bearer currencies to now our ledger-based electronic systems. Bitcoin and blockchain represent a transition to a new medium. This transition is often referred to as distributed ledger technology, which is a reference to today’s centralized ledgers. But I find it more helpful to look back to bearer instruments, like banknotes, to appreciate what this new medium enables: a digital bearer instrument. It’s so fitting that we are surrounded in this room today by antique U.S. banknotes, which are bearer instruments.
A key principle of a bearer instrument is that control of an asset equates to owning that asset. If I hold this $20 bill in my hand, you know I am the owner. But if I hand it to you, you are now the owner by virtue of the fact that you now hold and control it.
So here’s a pop quiz: now that I have handed you the $20 bill, how do we go about clearing and settling the transaction? It’s a trick question of course, because the answer is that we don’t have to. With bearer instruments the payment is also the settlement. It is one step. This is a neat property of a bearer instrument. But if the bearer instrument is physical, we have to be in the same room to achieve this. However, with digital bearer instruments like Bitcoin, we can achieve a single payment = settlement step over distance. For example, watch as I send the Wikimedia foundation some bitcoin. Now that I’ve sent it, we are done. The funds moved over the internet. It is as if I handed them cash.
But wait, don’t we have a digital payments system today? Not really. In almost all cases we have a digital messaging system, and those messages trigger a series of steps across multiple organizations: record, clear, settle, reconcile, etc.
The goal of the blockchain industry is to collapse these steps into a single step, where payment is the settlement, just like with physical notes. This is what I mean by digital value transfer, which I sometimes like to call money-over-IP. Soon, the phrase “cross-border payment” will make about as much sense as “cross-border email.”
Digital transfer without copying
So what makes digital bearer instruments possible? How can we skip the clearing and settlement steps? It boils down to one simple idea: using cryptography to transfer a digital object without also copying it.
Consider email. If I send you a PDF attachment, the PDF doesn’t disappear from my computer and appear on yours. You get a copy. The same is true if I text you a photo: it doesn’t disappear from my camera roll once you receive it. But if I have $10, and I send it to you, it’s very important that I don’t have the $10 after!
In contrast to these examples, on a blockchain network, the assets live on the network, and every participant holds what is called a private key that allows them to control the assets they own on the network. Thus, the thing they “have” is not the asset (in the way you would “have” a file), but rather a key that allows them to control and transfer the funds they own on the network.
So digital assets live on the network, but are controlled by keys held by the owners. This cryptographic control is why I call them digital bearer instruments, even though there is in fact a shared ledger recording each transaction. New technologies always lack perfect analogies!
If it helps, you can think of a blockchain like a wall of safety deposit boxes in the sky. Each box has a unique number (the public address) and a slot in the front so others can put assets into it. The box owner has the key that matches the box address, which means only the owner can access and move their assets. A blockchain is thus a push model, just like cash or other bearer instruments. This is in contrast to systems that operate on an “authorize and pull” model, like card networks.
We can put many different types of assets on blockchain networks. This allows us to trade assets without central counterparties. More complex transactions like multi-asset trades are often referred to as “smart contracts” in the blockchain industry. Remember, no clearing or settlement, even with something like a trade.
How do financial assets get on the network?
You are likely wondering at this point — “OK, but how to the assets get on the network in the first place?” Thinking back to Bitcoin for a moment, we know bitcoins are generated by the Bitcoin network on a fixed schedule and as a reward to miners for processing transactions. But this model would not make sense for issuing, say, commercial paper or U.S. Treasuries, as these are issued for business or policy reasons. And as we talked about before, we would not want to issue these corporate or government instruments onto the Bitcoin network, which is not designed for these asset classes.
So how do the assets get on the network? We use the same cryptographic principles that enable us to transfer assets (private keys with corresponding public IDs) to digitally mint assets onto a network. The private keys are like the minting machine, and the corresponding pubic ID acts like a CUSIP, or unique identifier for the asset class. We can also embed metadata into the issued asset so that the holder knows what it is, who issued it, how to redeem it, and so on.
For example, consider this hypothetical network of financial institutions, and a hypothetical issuance of commercial paper for Ford, who wants to borrow in the capital markets. Ford’s issuing bank generates a unique private key for the new Ford asset class, and publishes the corresponding Asset ID so the rest of the network knows what the asset is. Ford’s bank then signs issuance transactions with the private key for the Ford asset to mint individual units of the asset. The issuing bank then sells these to buyers who fund the purchase with digital dollars (more on these in a second). The transfer of the paper for dollars is a single step, what we called automatic delivery against payment.
The Ford commercial paper is a native digital bearer instrument, because the only medium it exists in is the network.
But what about the dollars that were used to pay for the paper in the trade? Where did those come from?
One possibility is that they were issued onto the network by a financial institution, who may be holding “real” dollars in a trust off the network. In this case, network participants who want digital dollars deposit funds into this trust and receive corresponding digital dollars on a 1:1 basis issued by the bank. Participants can then use the digital dollars to fund transactions on the network. We call this a title model, because the digital dollars in this case are really claims on funds held off the network. Depending on the context, holders of these digital dollars will convert back to the underlying periodically. But the digital title dollars can also circulate and be held on balance sheets if holders are willing to accept the counterparty risk of the issuing bank.
A better model, ultimately, is central bank digital currency. This would mean that a central bank, like the Federal Reserve, would participate on the network and would digitally mint U.S. dollars onto it. Since the Fed is the legal issuer of dollars, this would make digital dollars a native digital asset. There would be no need to “convert” back to some underlying “real” dollar because these digital dollars would be real, just in a new medium. In the same way that dollars already exist today in multiple mediums — notes, coins, electronic reserves — we would treat these digital dollars as just dollars in a new medium, backed by the full faith and credit of the U.S. government.
Once we have central bank dollars on the network, consider how simple a commercial paper trade becomes. One entity owns and controls some Ford paper. Another owns and controls a balance of digital dollars. Both assets are native digital instruments. A trade is proposed. Both sides agree, and sign with the private keys which correspond to the assets they control on the network. The assets swap, and we’re done. We get real-time asset movement with no clearing, no settlement, robust security, and perfect clarity around where the asset is at any point in time.
This suggests an even more fundamental question. If a central bank can now issue digital currency, why stop at issuing this currency to banks? Why not consider issuing it to non-bank financial companies? Why not corporations? Why not individuals? After all, banknotes can be held by anyone. And since anyone can in principle hold a private key, they can hold digital banknotes. Soon, everyone will have a smartphone, and all smartphones will have secure elements that can hold a private key. That key becomes the new leather wallet.
Better still, we can improve the security for people holding onto digital currencies with a technique called “multi-signature.” Remember that wall of safety deposit boxes in the sky? What if the boxes needed multiple keys to open them? That way, a user can have a key on their phone, but also have one held by a service provider, which could be a bank, but could just as easily be a company like Facebook or Apple. And a third key can be held in a secure backup location. To move funds, we would require 2 keys to create a valid transaction, which secures against theft or loss of their smartphone. This is just one example of how digital currencies provide more robust security than traditional banknotes, and more flexibility than centralized systems.
What does the future hold for central bankers and regulators?
Many people are asking the wrong question when thinking about blockchain technology. That wrong question is “How can we use a blockchain to streamline our existing systems?”
The right question is “What role should we play in the emerging digital asset economy?”
I put together a cheat sheet to help you begin thinking about possible answers to that question, depending on what your role is.
There are five roles one can play on these next-generation financial networks. You can initiate and operate a network, issue assets onto it, hold assets for yourself or on behalf of others, create products and services that run on the network, or, lastly, observe it in a regulatory or auditor capacity.
As central bankers, I would urge you to think about role 2 and 5 in the context of the networks being deployed by financial firms. These firms are attempting to improve the functioning of capital market use cases like repo and commercial paper in the so-called “shadow banking system.” CFTC commissioner Christopher Giancarlo recently said: “At the heart of the financial crisis, perhaps the most critical element was the lack of visibility into the counterparty credit exposure of one major financial institution to another. Probably the most glaring omission that needed to be addressed was that lack of visibility, and here we are in 2016 and we still don’t have it.”
Well, blockchain technology now provides that visibility (as well as protections for privacy). We now have a tool to measure systemic leverage and counterparty exposure. We can monitor compliance in real-time. We can answer questions about collateral ownership and rehypothecation that were at root in the run on the system in 2007. And when central bank digital currency powers settlement, the system will operate with fewer counterparties. What’s more, policy makers would gain a new tool for influencing liquidity in the increasingly important capital markets that operate outside of depository institutions.
Central banks, and clubs of central banks like the B.I.S., should also consider deploying networks of their own to issue currencies in this new digital medium. At first, it will probably make sense to issue these currencies to financial institutions and other central banks, but over time they should consider allowing individuals and businesses to hold them directly. This may require new licenses for non-financial companies who would hold private keys, and thus help to custody digital assets for people and companies. As asset custody is decoupled from financial products and services, it will increase competition and spur innovation.
Ultimately, blockchain networks will lead to a safer and better payments system. And central bank digital currency will be the foundation of that system.
There are many questions and implications of such a future, like the impact it would have on credit creation in the economy, whether it could or should be used as a new tool for monetary policy, and how to protect the privacy of those who hold digital currencies. But the technology is here and the first movers will find that it grants their currency a comparative advantage as network effects get created around faster and more secure payment networks. This is why we see central banks around the world already beginning to ask and answer these questions through real technology pilots.
Public sector leadership is key
Governments often play a key role as inventors or funders at the infrastructure stage of a new innovation, like the internet itself, GPS, and the microprocessor. We are in the infrastructure stage of the deployment of blockchain technology, so I would encourage you to begin learning, building, and contributing as the world shifts to a new medium for money.