Split or You Like to Be Told How to Govern

I applaud the SEC in particular for their courage to propose NMS governance structure reform amid large Exchange groups holding tremendous voting power in the Equity Data Plans. I have no objection to the Commission’s proposal because the root cause of today’s equity market structure problems is the unclear delineation of rights. I consent that the transformation of the Exchanges into publicly owned companies and the emergence of Exchange Groups exacerbate the conflicts of interest issue. The Commission’s proposed Amendments to UTP CTA CQ Plans aimed to strike appropriate balance between respecting the autonomy of Exchanges and improving best practices to advance the NMS related governance structure. Despite securities Exchanges registered with the SEC as self-regulatory organizations (SROs) are quasi-regulatory authorities, but SROs do not enjoy quasi-governmental immunity. Thus, I am not sure if the SEC can force the large Exchange groups administratively to accept the proposal ‘as-is’. Large Exchange groups are likely to fight back. That being said, I have long been advocating for “split up of Exchanges’ listing/trading business from data business” to replace the distorted economy of scope with efficiency gains from fewer fights and more cooperation among trading venues and better economy of scale.


The spinoff of Securities Information Processor (SIP) business from the securities Exchanges is fair to everyone because it relies on market mechanism to determine a settlement price for rights that SROs may willingly give up. This saves the hassle to re-design and maintain a fair and reasonable revenue allocation formula for distributing plan revenues, as participants’ voting rights are also going to be reallocated under the SEC’s proposal. The Exchanges will have no excuse not to accept this “split” because “if” they do not even trust the market mechanism that they themselves operate in for a fair price, then how they will be qualified to tell others to trade on their platforms?!

Another key advantage of “spinning-off” SIP is that it avoids potential litigation fights that delay the implementation of some of the governance best practices proposed by the SEC. There would no longer be challenges to overcome dominate voting humps hold by the Exchanges. In turn, SIP under new management can accelerate decision-making on advancing the SIP’s performance and price competitiveness with Exchanges’ proprietary products. That being said, this is under the condition of mandating the use of time-lock encryption for both SIP’s consolidated data and Exchanges’ proprietary feeds.

As mentioned in my submitted comments and meeting with the SEC in Dec 2019, I think the interpretations of 17 CFR §242.603(a) is incomplete and requires clarification or appropriate updates. “Transmitting or releasing data no sooner than to a Network processor (SIP)” only describes one of the aspects of “fair and reasonable” and “not unreasonably discriminatory” principles required by Reg. NMS. It omitted the fact that market data is highly valuable (it reflects the price discovery created by exchanges) and it requires proper security protection. Hence, the secured delivery (in-motion and at-rest) and retrieval of data in a timely manner are equally important. Time-lock encryption is a method to encrypt data such that it can only be decrypted after a certain deadline has passed. The goal is to protect data from being decrypted prematurely. So, regardless of when data is transmitted, everyone including subscribers of Exchanges’ proprietary products would only be able to start decrypting when the clock strikes to allow data decryption at SIP.

I already obtained patent approval on using Time-Lock Cryptography (TLC) to address this market data and market access problem. Rest assured this is not another speedbump. More importantly, the patent protection is a one-and-for-all approach eliminating the issue (market will not fix the market because of SROs rivalry and rent seeking behaviors) identified by Eric Budish, Robin S. Lee, and John J. Shim’s empirical research.

The design of TLC depends on accuracy and precision requirements, complexity to calibrate any nuances, and the amount of computational resources to decrypt the data. Because we do not want to push the bottleneck to an arms-race of using high-performing computers to decrypt data, therefore, computational resources and the type of data contents must also be considered in the design of a reliable encryption scheme. In order to match up the race in decrypting/ decoding data, my approved patent invention enables trade activities to transform into musical piece representation and metrical trees that making it possible to leverage Lossy compression methods, such as MP3, to yield a substantially greater compression ratio (60% or more of the original stream) as compared to traditional techniques (only 5-20%) that exploit statistical redundancy, Huffman coding, or probability method to represent and compress market data. My methods reduce data storage and booster the efficiency in data distribution, while also enabling the replicate the depth-of-book (DOB) information. Consequently, we are able to best preserve the richness of contents, while making the tool fast, easy, secure and fit for solving the market data and market access problem.

This brings us back to the point of why “spinning-off” SIP is better than prescribing a particular governance structure and continuous arguments over voting rights. When SIP under new management, they can quickly adopt both the TLC and MP3 approaches to benefit the industry. Next, per my submitted comments to the SIP operating committee, Market Authorities and the overall industry must recognize that, because of the immutable latency issue, there has to be some gives and takes to optimize between processing speed and contents’ richness for SIP. I encourage the industry to consider the following suggested priorities:

  • First, address  the  speed  differentials between  the  market  data  feeds  provided  by  the  SIPs  and  the proprietary products sold by the exchanges in order to get the biggest bang for the buck.
  • Second, demand for the depth-of-book information (a replication of the relative strengths in bid/ask price and  steepness  of  the  price  curve in  real-time), so at least the content would be a bit more compatible  with the proprietary products sold by the exchanges, while minimizing drags of the SIP processing speed.
  • Third, pursuit market  structure  changes outside  of  the  SIP  that  will  make  odd  lots  become  true “outliers” (Instead of bluntly get rid of odd lots by changing the round lot size, one should ask what are the various reasons for the existence of odd lots. For example, the causes related to a lack of stock split, the segment’s specific needs in moving blocks under the impending dynamics between lit and dark venues, etc.) rather than the “norms”, and/or ask for a “delayed” odd lot trades and quotations statistics, so that experienced market participants may use reverse-engineering methods to “figure-out” or “project” how these odd lots would play out in sequence. This “alternative”, “compromise”, or “trade-off” is based on the condition of Exchanges willingness to adopt point (i) – making market data available securely in synchronized time.

Click here to see my full 20-pages submitted comments to the SEC. Last but not least, be careful of what you wish for if you like the SEC to tell you how to govern. Certain steps would likely create bureaucracy, adding unnecessary regulatory burden. If rules are too prescriptive, it would affect private practices. Guidelines and procedures may never be comprehensive enough and regulators should not be in the position to judge these private practices or step-in every time when there are disputes among stakeholders. It would be impossible to put the Genie back in the bottle when regulators’ interference powers grow and be used or abused for political purposes. “If” there might be dysfunctional party, let them “off-the-bus” by spinning off SIP.