A large team of officers and staff affiliated with the Federal Reserve (eight with the Board in DC, five with the FR Bank of New York, one with the FRB of Chicago) have collaborated on a new paper on distributed ledger technology in payments, clearing, and settlement.
Of necessity, given the complexity of the subject, the team begins with a loose definition of DLT. It involves “some combination of components including peer-to-peer networking, distributed data storage, and cryptography.” The subject of the paper is how such elements in such combinations could “change the way in which storage, recordkeeping, and transfer of a digital asset is done,” the gist of it is that if there is a revolution coming, it likely won’t work out the way the most enthusiastic revolutionaries expect.
Fintech and its Visionaries
The subject is of understandable interest for the Fed. Payments, clearing, and settlement (PCS) systems in the U.S. at present process 600 million transactions each day. Changes in how this is done may have a great impact on the health and even the essentials of the business plans of the financial mediating institutions that play a big role in PCS at present and over whom the Fed has oversight responsibilities. The intermediaries themselves (banks or broker dealers) have their own intermediaries, and those come in different flavors: securities settlement systems, central securities depositories, and central counterparties. This is a complicated system that has grown up via the forces of spontaneous order. To some visionaries (troublemakers?) part of the promise of the new technologies is that they can render this complexity unnecessary.
The visionaries see every imaginable payments-relevant entity as in possession of a node, a device running DLT software that maintains database records. The nodes are connected to each other, sharing and updating that information. Every entity shares this responsibility directly with every other, peer to peer, and thus non-hierarchical, eliminating what the report calls “traditional database architectures that operate with a central hub that acts as the single source of valid information and control.”
Much of the report seems designed to tell the visionaries, “whoa! It isn’t that simple! Complexity will survive the revolution!”
Closed or Open? Levels of Encryption?
One important distinction within the DLT world is that between closed and open systems. In an open system, any entity with the technical ability to participate can participate, in principle as easily as anyone with an internet connection can edit Wikipedia. But closed systems have membership criteria. Existing members may require that potential new members provide proof of credit worthiness or have the necessary licenses to conduct the businesses they contemplate. While visionaries like to talk as if one big happy open system is on its way, the DLT arrangements actually under consideration for adoption by the financial industry are generally envisioned as closed systems.
Other important distinctions arise when one contemplates the particulars of systems. For example, consider the ledger itself. It is foundational to DLT that there is one common ledger of transaction histories and asset ownership that is shared amongst all nodes. But it doesn’t follow that all members of the system are equal in this regard. It is possible, the report reminds us, “that some of the data on the ledger is encrypted so that only authorized participants can decrypt and read the underlying information.”
So the ideal of peer-to-peer, with everybody on the same page as everybody else, may not match reality even if we look solely at the technical aspects of the proposed DLT revolution. But there are other aspects to consider.
“[C]ertain types of legal entities appear necessary to carry out ordinary PCS activities even as some of their functions and operations may adjust with the advent of new technologies,” the report says. CCPs for derivatives, for example, exist to be the buyer to sellers and to be the seller to buyers. For them to do this, some legal entity is necessary, and that type of entity will be in essence a financial intermediary, a hub with spokes. So it is tough to get rid of hubs and spokes architecture, disappointingly mundane though that fact may be.
Similarly, “traditional financial assets such as securities are typically a liability of a legal entity, such as a government or a corporation,” and any workable DLT arrangement must distinguish among entities based on their legal nature and liabilities.
After adducing such considerations, and a good deal more, the Fed’s team comes to the point where it might be expected to offer its own prediction: will there be a DLT revolution and if so, what will it look like? Disappointingly, the team rather throws up its hands at that point and says that time will tell, “[I]t is too soon to predict what these changes may be …the challenges identified in this paper will provide clarity over time.”
Until a process has concluded, then, one can draw no conclusions.