If there was any doubt that “blockchain” is the flavor of the month, the US Federal Reserve has issued a research paper summarizing the potential uses for “distributed ledger technology”, AKA blockchains [1]. In particular, the Fed considers possible transformations to “payment, clearing, and settlement (PCS) processes”
Their interest is obvious: the Fed has a “core mission” to “foster the safety and efficiency of the payment system and to promote financial stability” (p. 3). The Fed favors “innovation” but not “disruption”, and has legal responsibility and power to supervise these critical systems.
The report is an authoritative discussion of just what kinds of innovations and disruptions are at issue, at least within the arena of “Fintech”. The Fed has considerable historical perspective on such innovations. Here the report touches on two aspects: processes and participants.
Current Fintech processes have evolved over centuries, and everything is there fro a reason. Any effort to “innovate” will need to cover all the bases, and provide solutions to all the problems solved by current processes.
The same cannot be said for the participants, including intermediaries and institutions that implement an alphabet soup of infrastructure. These entities are entrusted with record keeping and other critical operations. The report points out that these entities keep their own ledgers, and must organize “hub and spoke” networks to reconcile the records. Distributed ledger technology has the potential to vastly change this landscape, shrinking or eliminating the role of these intermediaries.
Uses of Blockchain DLT
The report sketches the main features of the technology that are most relevant here. These are:
- Entities can be connected on a peer-to-peer basis via nodes
- Participants in a DLT arrangements can be permitted to play different roles or functions
- Ownership of an asset can be stored on a ledger within the DLT arrangement
- DLT arrangements can use cryptography to facilitate PCS processes
- Transactions histories and current states of ownership can be distributed across the nodes of the DLT arrangement
- The protocol in a DLT arrangement can define the procedures necessary for the asset transfer process
- Application programming interfaces can improve usability of DLT arrangements
- Smart contracts may be used in DLT arrangements to automate certain transfers based on pre-specified events agreed to by counterparties to a transaction
- DLT implementation can be considered from a legal entities perspective in addition to a technology perspective
The report is careful to note that the “peer-to-peer” protocols and APIs might be done in an open or closed system. A Nakamotoan style blockchain is typically “open”, in that anyone with requisite technical resources can participate as a node. But many of the explorations in Fintech use closed networks, which enforce legal and contractual requirements for the participating nodes. The technology works similarly, but the legal and social structure and relationships is substantially different.
Similarly, ‘who can do what’ on the blockchain might be done in different ways. The classic Nakamoto blockchain lets any node do any operation, but the protocol might enforce rule about who can do what. In Fintech, this approach is used to limit who can, say, issue new assets, while everyone can buy and sell them. Again, the technology operates in the same basic way, but there is an additional level of policy laid on top of it.
The report discusses how the consensus process is a central asset for this application of blockchains. The Nakamotoan design is specifically designed to support a consensus view of transaction history which is tamper resistant. Fintech systems might augment the basic protocols with additional measures as well.
However, there remain important questions about latency and scalability. The Nakamoto-style consensus has significant latency and limited capacity. The Fintech industry needs many orders of magnitude lower latency and higher capacity than the public Bicooin blockchain currently offers.
What might be done with this technology
Initial explorations are naturally focused on integrating blockchain technology into existing systems. This incremental approach will not necessarily revolutionize Fintech, but might still be significant.
The report cites a long list of potential advantages,
- “Reduced complexity (especially in multiparty, cross-border transactions)
- Improved end-to-end processing speed and availability of assets and funds
- Decreased need for reconciliation across multiple recordkeeping infrastructures
- Increased transparency and immutability in transaction recordkeeping
- Improved network resiliency through distributed data management
- Reduced operational and financial risks”
(p. 17)
They consider four key use cases for DLT.
- Securities, commodities, and derivatives transactions
- Cross-border payments
- Financial Inclusion
- Information-sharing
All of these have been discussed from Nakamoto on down.
Implementing securities and other transactions with a blockchain could be faster and cheaper, and reduce back office costs (not to mention thousands of jobs). Cross border payments are notoriously expensive and opaque, even for the largest players. Blockchains could reduce the number of intermediaries and associated costs, assuming that trust can be maintained and regulations can be enforced.
“Financial inclusion” refers to the goal of providing access to financial services for everyone, e.g., via simple smart-phone based APIs. Blockchain based systems are particularly interesting for this case because they are peer-to-peer, an because they can, in theory, handle the small and even tiny transactions that characterize the activities of many “unbanked”.
The report discusses the value of “Information sharing”, envision a blockchain where a “supervisor” (that would be the Fed) “would receive transaction data as soon as it is broadcast to the network, which could help streamline regulatory compliance procedures and reduce costs.” (p. 20)
Open Issues
The report gives a fairly comprehensive and useful sketch of many of the key issues that need to be worked out (see section.5). Some are very familiar (scalability), others have been hiding in plain sight all along (cryptographic key management). Remember, we are talking about the global economy, with bazillions of dollars at stake: hand waving isn’t good enoug.
For Fintech specifically, the report points out that the cost-benefit analyses have not been done, and these must consider the potential use of other (non blockchain) technology as well. Putative cost savings will have to be paid for, and the economics of operating a large scale, highly reliable distributed ledger system must be demonstrated.
The report sketches some of the “financial design” issues, which, “echo debates from the development of book-entry securities and electronic payments over the past five decades” (p. 25). For example, “DLT may provide opportunities to re-visit the traditional choices for the design, holding, and transfer of securities”, perhaps with creative uses of digital “tokens”. Similarly, banks may develop new, streamlined methods for managing monetary payments.
In addition, the role and even existence of intermediaries is up for question and redesign. Even if a blockchain based system might replace some of the functions of an intermediary, they have other functions (such as match making), which will presumably still be needed.
The report also gives a useful summary of some important “risks”, including legal and regulatory (see section 6). In addition, there are substantial financial risks for all parties that need to be defined, including the definitions of defaults, errors, theft, security, and operational failures, and who is responsible.
For starters, what is the legal status of a record on a blockchain? That depends on where it came from and how it got there. In the case of documents that must have legal authority, there must be adequate protocols to assure that the record on the blockchain is authoritative. Ditto for all forms of record: you need to be able to trust the information.
Not surprisingly, the Fed is crucially interested in maintaining adequate governance, including enforcement of secrecy and money laundering controls required by current law.
As the report says, “such risks are typically concentrated at intermediaries such as banks and FMIs, which, in turn, specialize in managing and controlling these risks.” (p. 27). Innovations that are meant to increase efficiency and mitigate some risks may well actually just shift the risk among the parties. The overall impact of proposed systems need to be considered, and there will likely be serious cost-benefit decisions to be made.
“For example, on the one hand, if a solution reduces the cost of clearing, but does so by shifting financial and operational risk to end-users of a system or by reducing the ability of specialized intermediaries to manage risk, difficult decisions may be needed about the trade-off between the benefits of greater efficiency and the associated costs and risks.” (p. 27)
Where the Fed is Skeptical
Much of the report is a fairly conventional review of blockchain technology and its potential uses. There are several points on which the report is strikingly skeptical, which we should pay careful attention to.
One thing that will not change is the need for trust. The financial system is largely about “trust”, so any innovations must maintain that trust.
“Many models may alter or eliminate some roles of current intermediaries in payments, clearing, and settlement but may not necessarily eliminate the need for coordination or centralization of certain functions by trusted intermediaries.” (p. 16)
It’s fine to think about eliminating the costs associated with intermediaries, but that cannot be done at the expense of the trusted relationships they provide.
It is notable that the report gives a very cool reception to “smart contracts” and distribute autonomous organizations. This is quite understandable given the dubious design and history of these devices to date. To the degree that the Fed is concerned with the stability and soundness of the financial system, it will continue to be leery of robotic mechanisms possessing neither brakes nor steering wheels.
In addition, the very notion of an organization with no human in charge is shear nonsense within the world of Fintech, which is all about accountability.
“If “management” of an organization is conducted automatically by code, legal systems will have to determine who to hold accountable if laws are broken and disputes arise.” (p.29)
Throughout the report, the authors make the point that “information technology is only one factor”, Fintech systems are also driven by business needs, network economics, and legal issues. Blockchain enthusiasts may be excited by the technological possibilities, but actual deployment will be shaped as much by these non-technical forces. In short,
“A key challenge is identifying appropriate use cases where the potential reduction in costs of operational and financial inefficiencies would justify the cost of the investment and operational changes needed to implement DLT.” (p. 22)
Overall, this is a useful and authoritative discussion, from folks who have a lot of experience with financial technology, and who have a remit to steward the integrity and stability of the financial system.
- David Mills, Kathy Wang, Brendan Malone, Anjana Ravi, Jeff Marquardt, Clinton Chen, Anton Badev, Timothy Brezinsk, Linda Fahy, Kimberley Liao, Vanessa Kargenian, Max Ellithorpe, Wendy Ng, and Maria Baird, Distributed ledger technology in payments, clearing, and settlement. Board of Governors of the Federal Reserve System, 2016-09, Washington DC, 2016. https://doi.org/10.17016/FEDS.2016.095
Cryptocurrency Thursday
Robert McGrath's Blog 