The Financial Stability Oversight Council (FSOC) released its annual report on the health of the U.S. financial system on June 21, 2016. I believe it diagnosed many issues correctly, but missed an opportunity to encourage a key innovation that would give the FSOC what it repeatedly said it needs in the report: better data.
That innovation? It’s called “blockchain” or “distributed ledger technology.”
I write this post as a 22-year Wall Street veteran who’s seen a lot. I’ve worked directly for the co-CEOs of a global bank and had signing responsibility for the financial statements of regulated subsidiaries at various points. So I understand the issues on FSOC’s plate and respect the difficulty of its task. Oh, and I’m a blockchain “veteran” too—as much as that’s possible with nascent technology!
The FSOC report repeatedly called for better and faster disclosure of information that could enable regulators to assess systemic risk more effectively. Yet, instead of recommending that FSOC members explore whether blockchain technology can answer these calls for better data (as their brethren in the UK, Europe, China and Japan have done), the FSOC relegated blockchains to the second to the last page of the report (page 127 of 128).
The report’s discussion of blockchains highlighted that they may enable transactions to occur without trusted third parties, may “significantly improve efficiency by replacing manually intensive reconciliation processes and reduce risks associated with trading, clearing, settlement, and custody services.” FSOC then focused on the risks that participants and regulators would need to monitor, including operational vulnerabilities “that may not become apparent until they are deployed at scale,” and to the potential for “fraud executed through collusion among a significant fraction of participants in the system.”
But the FSOC report did not mention the possibility that blockchains can provide the improved disclosure for which its report made repeated calls.
In this post I hope to help regulators understand why blockchains can achieve this, and more. I wrote this post primarily for an experienced audience in the financial services regulatory field. For those not familiar with the jargon, the FSOC report has helpful background and a glossary (available here). As always, I provide no investment advice on this blog and you should refer to important disclosures here.
BACKGROUND ABOUT THE FSOC REPORT
For readers not familiar with FSOC, the Dodd-Frank legislation created it (1) to identify risks to financial stability of the U.S., (2) to promote market discipline by eliminating expectations that the U.S. government will bail anyone out in the event of failure, and (3) to respond to emerging threats. FSOC is an inter-agency council meant to coordinate efforts of the many different regulators of the U.S. financial system, including the U.S. Treasury, the Federal Reserve, OCC, SEC, FDIC, CFTC, state insurance regulators and other agencies.
The report, in addition to assessing risks in the financial system, among other things, is required to make recommendations “to enhance the integrity, efficiency, competitiveness, and stability of United States financial markets, to promote market discipline, and to maintain investor confidence.”
FSOC’S REPORT SUMMARIZED IN JUST THREE WORDS: “NEED BETTER DATA”
The executive summary of the FSOC report called 10 separate times for better access to information about the financial sector, citing the need for “improved collection of data,” “enhanced reporting and disclosure,” “improved transparency,” and more work to resolve “data gaps.” The executive summary then addressed developments in distributed ledger technology (a.k.a. blockchain), without mentioning the possibility that blockchain could solve all of these specific needs for better information.
ISSUES IDENTIFIED BY FSOC AND POTENTIAL BLOCKCHAIN SOLUTIONS
I’ll examine issues in the order listed by the FSOC report—with the exception of cybersecurity, which I’ll handle last. FSOC listed cybersecurity first, but I’ll address it last—truly saving the best (but most dense) discussion for last. Otherwise this blog post mirrors the FSOC report’s numbering, which begins at section 3.2 for those following the report side-by-side.
3.2 Risks Associated with Asset Management Products and Activities
In my opinion the FSOC got this right, properly identifying this as the top risk to the financial system after cybersecurity. I haven’t been afraid to criticize the financial industry by admitting no one really knows how leveraged the financial system is, but I’ve also defended the industry against its critics by affirming that the industry wants improved visibility into systemic leverage more than anyone else. So why hasn’t this been fixed? Answer: it’s an information problem. Let’s dig into FSOC’s specific assessment, and I’ll describe how blockchain can solve many of these information problems.
FSOC broke this topic into (i) liquidity and redemption, (ii) leverage, (iii) operational functions, (iv) securities lending and (v) resolvability and transition planning.
3.3 Capital, Liquidity and Resolution
FSOC notes that depository institutions have meaningfully increased capital levels and liquidity buffers, and I applaud these moves. The financial system is still too leveraged, in my opinion, and outside of the FSOC report it seems regulators still agree this is the case. The large stock price reactions of the biggest banks to the Brexit vote would seem to support that view, as does the fact that in April the Fed and FDIC notified five of eight of the largest, most complex bank holding companies that their so-called “living wills” were deficient.
The FSOC report recommended that the Fed finalize rules for minimum capital requirements for the U.S.’s global systemically important banks (U.S. G-SIBs) and large foreign banks (FBOs) operating in the United States, and that the agencies “take appropriate action” to handle the living will statuses of large banks.
Again, blockchains could provide real-time transparency into the true risks of all financial institutions, and this would facilitate resolution of failed depository institutions. Tracking of risk across multiple legal entities and jurisdictions is a challenge even for the banks’ risk departments themselves, much less regulators, and the “living will” regulatory process gives banks clear cost and capital incentives to deploy IT solutions that can transparently aggregate risk reporting in real-time.
3.4 Central Counterparties
In this section, FSOC maintains that central counterparties (CCPs) are “key to the effective functioning of a number of markets.” Here, I would respectfully question whether this is always true—especially for cases in which the CCP adds counterparty risk where it would not otherwise exist. A blockchain would obviate the need for central counterparties to use their balance sheets, because there would be no need for clearing in a world in which trades settle instantaneously (delivery-versus-payment). CCPs would not necessarily disappear in a blockchain world—they would probably still provide value-added services, but client assets would no longer touch their balance sheets and therefore their systemic risk would be much, much smaller. Here’s how I would conceive of possible services CCPs could provide in a blockchain world: administering the blockchain, setting its standards, handling permissioning/entry and exit of participants, acting as arbiter of disputes, providing big data analytics, and much more.
3.5 Reforms of Wholesale Funding Markets
3.6 Reforms Relating to Reference Rates
To address weaknesses in certain key market benchmarks—especially the continued reliance on LIBOR—the FSOC report recommended that Alternative Reference Rates Committee (ARRC) identify alternative, near risk-free rates. FSOC noted the goal of well-governed benchmarks that are “anchored in observable transactions and resilient against attempted manipulation.” Blockchains can assist in achieving both goals, since the transactions and prices would be transparent to regulators. Blockchains are especially useful for creating an agreed source of reference data, precisely because transactions are transparent and immutable—so data pertaining to these transactions would be readable directly from an easily auditable ledger rather than self-reported (sometimes subjectively so). I believe market participants would welcome having an immutable transaction history upon which to base such critical benchmarks and analytics.
3.7 Data Quality, Collection and Sharing
Nowhere else can blockchains help regulators more than in the areas of data quality, collection and sharing. I’m broken record on this point!
This section of FSOC’s report began, “Addressing data needs for the analysis of potential threats to financial stability remains an important priority of the Council…” It then broke the section into several sub-sections.
3.8 Housing Finance Reform
The FSOC report understandably focused on Fannie Mae and Freddie Mac (the GSEs), and FSOC reaffirmed its “view that housing finance reform legislation is needed to create a more sustainable system” and to take steps “to encourage private capital to play a larger role in the housing finance system.” The FSOC report didn’t highlight the area of mortgage finance in which blockchains can really help: improving recordkeeping and servicing of mortgage loans. In a blockchain world, deeds, mortgages and liens would be recorded and transferred via a blockchain, not in paper form. This “dematerialization” of records would facilitate securitization issuance and servicing, as investors would have real-time transparency into the cash flows of the individual mortgages that back securitizations—instead of waiting for monthly remittance reports. Disputes would be resolved more quickly. The foreclosure process would be smoother. Mortgage servicing would still have a human element, of course, but would be fairer and more transparent to both sides.
While full conversion to a blockchain world will take years, I see the Delaware Blockchain Initiative as an important first step to dematerialization of corporate registrations, liens, mortgages, UCC filings, deeds, and much of the paper involved in securities issuance.
Something Else Worth Discussing
I noticed something later in FSOC’s report worth discussing, which pertains to the drying up of bond market liquidity. FSOC alluded to the issue in Section 3.10, looking at the impact of market structure changes on Treasury bond market liquidity specifically—but liquidity has dried up across the broader bond market too.
There are many reasons for this, likely the biggest of which is higher capital requirements that have caused securities dealers to reduce the inventory of bonds they hold to facilitate customer flows.
When reading page 65 of the FSOC report, I drew an analogy to the bond market liquidity issue. On page 65 FSOC discussed the improvement in swap market liquidity since 2014, when swap execution facilities (SEFs) came into the picture. The analogy between swaps and bonds is far from perfect, of course, but the bond and swap markets shared something until 2014 (when the swap market began migrating to SEFs): they were both over-the-counter (OTC) markets. Two characteristics of OTC markets, in my experience, inherently limit their liquidity—(1) opacity in price discovery, and (2) participants assume the risk of settlement failure between the moment of trade execution and settlement (which, for securities, is T+3 days).
On page 65, FSOC examined the improvement in swap market liquidity when it moved to SEFs, which improved price transparency and competition and reduced counterparty risk. FSOC said, “A recent study by the Bank of England finds a positive link between SEF trading (both voluntary and mandatory) and a significant improvement in liquidity, in particular for USD-denominated interest rate swaps which are most affected by the SEF mandate. The study postulates that the increases in volume and market liquidity result from enhanced transparency and the reduced search costs provided by the SEFs. The associated reduction in execution costs associated with SEF trading is economically significant.”
Blockchains, in my opinion, would boost bond market liquidity for the same two reasons SEFs did for swaps. Blockchains could reduce or eliminate counterparty risk (by speeding up settlement, potentially to the exact moment of execution (delivery-versus-payment)—thus obviating the need for assets to travel through the balance sheet of an intermediary, such as an SEF, exchange or other central clearinghouse, before settlement). And further, blockchains could facilitate electronic bond trading at the pre-trade juncture, just as SEFs did for swaps—helping price discovery by enabling participants to send anonymous request-for-quote queries to owners of bonds (to facilitate the search for buyers and sellers), and possibly even incorporating matching engines into the blockchain itself.
I believe withdrawal of dealer liquidity is the spark that finally makes the bond market ready for electronic trading, which has had several fits and starts but hasn’t happened yet because dealer inventories had kept markets functioning. It’s fortuitous that the advent of blockchain technology is happening just as dealer liquidity has dried up.
No other market is more poised for transition to an institutional peer-to-peer, dematerialized (i.e., not paper), electronic model than the bond market, in my opinion. Most of the bonds that will be outstanding in 10 years haven’t been issued yet, and so the bond market could adopt blockchains for new issues only without retrofitting legacy infrastructure for securities already issued.
New bonds could be issued directly onto a blockchain and live natively there—existing in dematerialized form from inception to maturity, with everything in between (trading, payments of interest and principal, calls, tender offers, exchanges, securitization, remittance reporting, etc.) happening directly on a blockchain. Again, the Delaware Blockchain Initiative is a critical step in making this legally possible, as corporate actions, mortgages, liens, UCC filings, and other items should soon be able to be recognized legally on a blockchain (in lieu of paper). Some these options should be production-ready later this year, with more to follow in 2017. I would encourage FSOC to dig into what this means, and am happy to engage in a direct dialogue!
I’ll omit discussion of the last two recommendation sections of the FSOC report (3.9 “Risk Management in an Environment of Low Interest Rates and Rising Asset Price Volatility” and 3.10 “Changes in Financial Market Structure and Implications for Financial Stability”) since I don’t see much application for blockchains to either topic.
3.1 Cybersecurity (saving the best for last!)
Finally, let’s turn back to cybersecurity, the top issue in the FSOC report. I agree it should be FSOC’s #1 priority. Yet I’d also rank cybersecurity as blockchain’s best attribute. Owing to the technical nature of this discussion, which may lose some readers, I elected to handle it last.
The laws of math have proven a lot more tamper-proof than the laws of man!
Blockchain, as an IT security model, is superior relative to the model of current financial infrastructure. On a blockchain data is decentralized, can be fully encrypted, and can be used as the statement of record for all state changes in the lifecycle of an asset. Additionally, data is replicated throughout a network, making it far more resilient and making hacking attacks far more unlikely and costly. By contrast, today most data is fragmented, inaccessible, unencrypted and stored behind perimeters that have proven too easy to penetrate, unfortunately. Fragmented architectures inherently have attack surfaces that are wide and multiple points of failure that are too vulnerable to attack.
To understand the four corners of what’s possible, I would encourage FSOC to study the superior IT security attributes of blockchain’s most visible standard, bitcoin. Please, set aside what you’ve read about bitcoin and focus specifically on its design as IT architecture. It has a great deal to teach about cybersecurity.
Bitcoin’s blockchain has survived for 7 years, completely open and exposed to the 24/7/365 war zone of Internet security, without a single successful attack on its ledger. Many sophisticated adversaries have tried unsuccessfully to steal its hacker’s bounty, which is now worth about $10 billion. So far, attackers have succeeded only in hacking applications built on top of its protocol, rather than the data itself.
It’s informative to dig into why Bitcoin’s cybersecurity has proven so strong. The magnitude of computational power required to break the Bitcoin blockchain—or even to re-write the history of this morning—is so staggering that it’s unlikely to happen. And bitcoin visibly grows more robust nearly every day. You can track the growing computational power of its network here, which as of this writing stands at 19,205,707.43 PetaFLOPS (a measure of computer performance). This towers over the world’s fastest supercomputer, China’s new Sunway TaihuLight, which runs at a comparatively miniscule 93 PetaFLOPS.
A critical, intertwined design feature reinforces bitcoin’s defenses against hackers. It’s called “difficulty,” which is how hard it is to perform the calculations required to add a block to Bitcoin’s blockchain. Bitcoin’s protocol automatically adjusts its “difficulty” to keep block calculation times to ten minutes, so as more computer power (hashpower) comes into Bitcoin’s network, its difficulty rises.
Because of this “difficulty” feature, even if a bad actor were able to gather up enough computational power to attack the Bitcoin network (including a so-called “51% attack”), it would not succeed in altering more than a few blocks on Bitcoin’s blockchain (i.e., a few hours of its history). The amount of computational power required to steal bitcoins from the blocks added yesterday, much less those of last month or last year, is pretty much unfathomable.
It is precisely because of this lopsided upside/downside calculus—requiring an awesome amount of cost and computing power to gain so little—that no hacker has succeeded in breaking the Bitcoin blockchain.
And then there’s this. As of this writing, 5,730 nodes keep a watchful eye over the bitcoin network, located all around the world. From a cybersecurity perspective, this is powerful—all nodes have strong financial incentives to protect the network from attack, and all nodes maintain a complete backup copy. Each node is itself a complete business continuity plan.
Can you think of a more secure IT system than what I’ve just described? I cannot. Can any bank say that the financial reward from a successful attack on its infrastructure is small? Recent history has unfortunately proven otherwise.
To understand the cybersecurity benefits of blockchains, whose IT ancestry is bitcoin, I would encourage FSOC to study why bitcoin has proven so resistant to cyberattack, and whether any—or all—of its attributes might be applied to improve the cybersecurity of the U.S. financial system. Most of the financial institutions regulated by FSOC are studying this very question, and rightfully so.
The FSOC report correctly identified the most critical and pressing information gaps that pose systemic risk to the financial system. No one really knows how leveraged the financial system is. Yet, the report missed connecting the dots about the new tool that might finally give them the information they need to assess that very risk: blockchains.
I recognize and respect, more than most, the near-impossible job of financial regulators in tracking the highly complex and inter-connected global financial system. Regulators have made great strides since the financial crisis. But I believe the system is still too leveraged and opaque. There’s more work to do.
I reiterate my open call to senior regulators to help educate you about the potential of blockchain technology to benefit all participants in the financial ecosystem. Blockchains are not easily grasped in one setting, and I encourage all regulatory influencers to dive in and engage with us in the industry! We have more common ground than you’d think!
Bitcoin/blockchain, ex-Wyoming Blockchain Task Force, 22-year Wall Street veteran.
This web site is limited to the dissemination of general information, investment-related information, publications, and links.
Please consult important additional information and qualifications HERE.