The US Court of Appeals struck down the SEC’s CT-Plan on July 5, 2022. Without the CT-Plan to “outsource” certain SEC authority to the operating committee (2/3 SROs and 1/3 Non-SROs), how will Market Data Reform likely evolve? The Self-Regulatory Organizations (SROs), including FINRA, have stated their willingness to provide part of the expanded data under the Market Data Infrastructure Rules (MDIR). Their Odd-Lot proposal for the Securities Information Processor (SIP) may be a curve ball to the SEC. How should the SEC and market participants react or respond?
As a quick recap, the SEC’s MDIR expected that Competing Consolidators (CCs) under a Decentralized Consolidation Model (DCM) would “permit the market data infrastructure to more readily adapt to changes in technology to better fit the needs of market participants…” Many welcomed the MDIR to expand contents of core data (including depth-of-book, auction information, and more). The SROs obviously disagreed to give away the expanded content for FREE, so everyone would likely be getting a piece of the cost.
The tricky part is whether Odd-Lot Quotations under the SIP proposal, and/or other expanded content under the MDIR, will be disseminated through a single, integrated feed, or through a segregated feed providing subscribers choice? Ingesting a lot of data raises the concern about bandwidth connectivity. Lacking depth-of-book data undermines the usefulness of Odd-Lot data. Nevertheless, compilation of protected quotes is complicated. Citadel Securities did raise legitimate questions in their comment letter. What you see may not be what you get. One will need to upgrade to higher bandwidths and add depth-of-book data, or else face being disadvantaged to prop feed subscribers.
The SEC initially left some of the above to be determined by the CT-Plan operating committee and set a tight deadline for MDIR implementation. Now they will have to provide definitive answers that may face further legal challenges. With the Court ruling overturned the SEC’s decision to reject fee-increases for Exchanges data feeds, the SEC does not necessarily have the upper hand against the Exchanges’ power-play. The SEC also did not hold back IEX and MEMX from charging fees on real-time market data, effective July 1, 2022, which used to be FREE. Everyone is for themselves in the Animal Farm. Connecting to more venues for BestEx compliance is an additional cost to market participants. Demand for the Exchanges’ proprietary products (PP) is inelastic. Empirical Researchhas shown that Exchanges may optimally restrict access to price information by charging a high fee so that only a fraction of speculators buy their PP. The Market will not fix the Market because of rent seeking behaviors.
The assumption of the SEC’s MDIR that based on 10G speed is fundamentally flawed. The SEC let the NYSE to implement 100G connections for their proprietary feed in April 2020. Widening the latency gap with overly frequent upgrades and exploiting any residual data disparity exacerbate inequalities between the “haves” and “have-nots”. I listened to the SEC DERA event on May 6, 2022 – the Market Microstructure Track, particularly “The Market inside the Market: Odd-lot Quotes” session. It was useless and disheartened. It should not require a Machine Learning (ML) model to affirm there is value in Odd-Lot data, no one is denying that. The problem is – average investors lack the speed to access and may not hold the power of ML to compete. Denying the need to enforce a Speed Limit is either naïve or disrespect to the wisdom of former SEC Chair Mary Jo White who famously said, “deemphasize speed as a key to trading success…”
What problem is MDIR trying to solve? Right from the start, it has always been about aggregation distance/ location differential issues. “Same manner and methods” may merely be a standard price list offered by Exchanges under the MDIR. Neither is it equivalent to Latency Equalization, nor can it achieve the same results as using time-lock encryption to make Market data available Securely in Synchronized time. The SIP significantly underperforms as compared to PP sold by ‘for-profit’ stock exchanges with monopoly power. The bigger problem for Market Date Reform is indeed “who owns the data’. While elites are getting 32 mils super tier rebate and other privileges, others may get nothing.
Would CCs without the CT-Plan’s “governance oversight” be able to spread their fixed cost across a larger base of consumers (in benefiting the industry to strike for a “fairer, reasonable and non-discriminatory” (FRAND) outcome? The reality is that market data costs are increasing rather than decreasing. When policy making and market reform is delegated or left to the hands of a few, the implicit risks of cahoots would be higher. Bureaucracy may be created to contest for the party’s interest rather than the public’s interest. Further, how the divergence between private and social costs would be assessed is questionable. In turn, non-contracted parties and underrepresented groups will suffer. I cannot comprehend former Director Redfearn’s tweet that said “This is not a loss for the SEC and market participants. This is more like a double or a triple instead of a home run…” What’s the point of salvaging one’s pride, when the last SEC administration clearly majored in the minors!
Let’s be honest, which market data vendors would be able to make a difference if they become CCs? Maximizing the life of aged technologies beyond its 10 years amortization period is in the large established firms’ best self-interest. The latency gap between software based streaming versus Field-Programmable Gate Array (FPGA) processing is huge. Exergy (merged with Vela) and NovaSparks have Patents in FPGA. NovaSparks was acquired by Peninsular Capital in a Management Buy-Out (MBO) transaction in 2020. MayStreet got $21 million in their Series A in 2020 and then another $10 million in debt financing in 2021, and most recently they have been acquiredby London Stock Exchange Group (LSEG) that also own Refinitiv. Please correct me if I am wrong, MayStreet partnered with NapaTech for FPGA, so I do not know whether LSEG would buy the FPGA component next.
All in all, if any of these ‘display’ or ‘non-display’ vendors would have made a difference, they would have succeeded long ago. They together with many self-aggregators, Alternative Trading Systems (ATSs), Smart Order Routers (SORs), Transaction Cost Analyzers (TCA) and other Best Execution (BestEx) tools are indeed added layers of intermediary costs to the industry’s value chain. Their existence serves to fabricate the fragmented markets like bandages. There is no incentive for them to truly address the location differential issues and data ownership problems because that contradicts with their business models. Crowning them with CC licenses may imply authorized rights for them to rent seek on data subscribers.
Realize the gap between MDIR and where market data and market structure reforms should go. What the current SEC administration should do is reject the SIP Odd-Lot proposal and hold off any market data feed upgrade or fee increase. What the market needs is a solution provider whom understand time-lock encryption; a good data modeler who knows about consistency, performance degradation, B trees/ LSM/ Fractal tress, careful of what being parsed in-memory, read/write, partitioning and whatnot to tune a lossless or lossy feed faster than “FIX Fast” format, assuming all FPGAs being equal; as well as someone who has the business sense to contrive a new mechanism to replace the skewed rebates, payment for order flow and other distorted incentives that weed out conflicts of interest and promote FRAND. We, at Data Boiler know enough of the above to discern what works and what does not. Next, we will talk about “order-by-order competition” (routing retail stock orders to auctions) and an alternative, stay tuned!