Noumenon of Equity Market Structure 

By Kelvin To, Founder and President, Data Boiler Technologies

Noumenon of US Equity Market Structure to some extent resemblances the parable of “blind people and an elephant”. Different people may use different phenomena that they experienced to describe the big picture differently. Yet, Laozi’s Taoism said, “A semblance great, the shadow of a shade” (big picture is formless). So, people have debated about market designs, continuous trading versus auction, venue-by-venue versus order-by-order competition, tick size, BestEx, odd lot, round lot, and whatnot over-and-over again for decades. Using takeaways from recent SIFMA roundtable, the enclosed whitepaper analyzes what is the noumenon of National Market System (NMS) and where it should or is heading. The following are some of the highlights and key points:

A semblance great – the US equity market being envy of the World. Yet, complacency may lead us to forget the noumenon of Regulation NMS, or the so-called “venue-by-venue competition” is indeed a brutal Warring States Period.

  • Reg. NMS is a departure from one-size fits all. Vigorous efforts were made to fabricate the highly fragmented US equity market to bear fruit for the economy. These prominent players at the SIFMA roundtable may feel underappreciated, yet they are indeed the pioneers of decentralized finance (DeFi), not the Blockchain DLT folks. 
  • From the regulators’ perspective, governing CeFi is straight forward. As markets evolved from CeFi to DeFi, the Commission adjusts and shares governance responsibilities with the SROs. The only way to regulate the highly fragmented equity market (DeFi) is via decentralized crowd intelligence.
  • For all fairness, DeFi does not have absolute advantage over CeFi, and vice versa. CeFi is superbly efficient up to the limit of ‘big brother’ deciding who gets what. Yet, ‘big brother’ does not always make the best decision and can be subjective, causing favoritism and possible corruption. Whereas DeFi replaces the ‘big brother’ with a network that consists of bilateral and/or multilateral agreements that are sometimes referred to as bureaucracy. 
  • Main Street’s skepticism of Wall Street’s bureaucracy is understandable. Rogue trading has caused chaos in the past and may still will. We doubt the effectiveness to rely on “fire and replace the execution vendor” approach to hold the market makers’ feet to the fire. Controls may be weakened or bypassed through transfers in-and-out of category between available-for-sale and hold-till-maturity and/or a flipping-switch between dealing with “client” versus “counterparty”. We do not buy the argument of alluding to the problem as a principal agent issue. Public confidence has been eroded by a series of cases related to the order routing enigma and blindsided of risks.

Mercy to the ordinary market participants (the smaller players in both institutional and retail), they jump through hoops and are subservient to the elites to stay in business. The gap between the ‘haves’ and ‘have-not’ is too wide.

  • Trading venue perimeter allows for assorted incentives and business models. Elites are getting 32 mils super tier rebates and other privileges, while others may get nothing.
  • The assumption of the SEC’s market data infrastructure rule (MDIR) that is based on 10G speed is fundamentally flawed. The SEC allowed the NYSE to implement 100G connections for their proprietary feed in April 2020. Denying the need to enforce a Speed Limit is either naïve or disrespect to the wisdom of former SEC Chair Mary Jo White who notably said, deemphasize speed as a key to trading success”.  We advocate the use of time-lock encryption (TLE) to make market data available securely in synchronized time. Rest assured that TLE is not another speed bump.

Contemplating tick size/ minimum pricing increment, redefining NBBO, accelerating odd-lot/ round lot, and favoring auctions could well be a separate discussion about different market centers’ capabilities.

  • We concur that “when the number of price points is increased, then it essentially fragmenting the quote and dispersing liquidity across multiple prices. It is a balancing act between the high price and the low price securities when considering changes to tick size regime.”
  • We share the concern about the retail experience if the definition of NBBO is conflated. We also understand the institutional prospective. They are focusing on slippage versus NBBO and not price improvement with the NBBO. Points we would like to add about odd lot proposal for SIP: (1) Ingesting a lot of data raises the concern about bandwidth connectivity. Lacking depth-of-book data undermines the usefulness of Odd Lot data; (2) eroding the NBBO de facto status is like the transition from LIBOR to SOFR. 
  • Hype about routing retail orders to auctions generated more questions than anything. It would likely run into conflict with other rules, as well as raises fairness/ anti-competitive issues. Reference to 79 FR 5592, footnote 711, it said “The Agencies are not adopting a ‘transaction-by-transaction’ approach because the Agencies are concerned that such an approach would be unduly burdensome or impractical and inconsistent …” This cannot be more confusing regarding the Commission’s policy direction. 
  • Exchanges’ retail liquidity program (RLP) can only be complimentary, but not competitive with off-Exchange markets. We found the frienemy relationships between Exchanges and Internalizers amusing.
  • Disclosure rule was developed in 2000 and people are still questioning for whom this information prepared. Metrics are rarely effective to deal with rapidly evolving issues proliferated by hidden problems and silos. Wholesale block trading is highly situational. BestEx for “where” customers’ orders are routed versus “when” routing should happen are complex matters, even to some of the participants at the SIFMA roundtable. We suggest replacing the current 605 reporting with some sort of annual inspection sticker mechanism.

The shadow of a shade – Access fee rebate, PFOF, and market data/ market structure issues are all intertwined. At Data Boiler, we believe What Gets Paid and Who Gets What must base on clear delineation of rights and obligations. The noumenon of rebate incentives serve as royalty payment for the use of others’ copyrighted material. Think about what gives rise to arbitrage or pick off on price. Anyone would have done it if they do not have to bear the corresponding cost in using others’ copyrighted materials. BestEx, order protection and ways to curb conflicts of interest all boiled down to tweaking incentives causing it to be economically not viable for a constituent to exploit its economy of scope, scale, or engage in other misbehavior acts. 

The US equity market never lacks competition. Indeed, we have more than enough markets but insufficient “Farmers” (see this value chain smile curve) and diversity to work in the field to bear fruit. Market operators make more money selling “opioids” (TCA, liquidity sourcing, outsourced execution tools to fabricate fragmented markets are like addictive drugs) than “fruit”. Warlords are staking claims everywhere in DeFi division of the markets, if the SEC does not want to be perceived as price control interference of markets, the only way to drive orderly function of markets is by aligning and making values transferable across different trading (streaming) platforms. 

For that, copyright licensing mechanism from the music industry has shown us a roadmap. Data ownership/ intellectual property right is the only thing that can address the issues across the access fee rebates, PFOF, market data and whatnot. Rest assured it is market driven, and not subjective judgement on what should be the equivalent value of price versus size improvement. 

  • The approach is simple – trading platforms would be based on who they want to serve, how many subscriptions they are going to get and determine whether to carry a broader or narrower “catalog”. The broader the “catalog”, the platform would pay a wider range of broker-dealers, featured traders, algo developers in royalties. 
  • Best of all, it is time tested against decades of lawsuits in other industries to assert data ownership claims, as well as effective to deal with bilateral and multilateral agreements embedded in formal and informal networks.
  • This copyright mechanism and our suggestion of a “sound library” would allow for creative discovery of unknown unknowns, which can be anybody’s game. It will minimize the gaps between the ‘haves’ and have-not.
  • There will be opportunities to earn royalty’s payments, 2nd profit for unused algorithms, off-loading of certain staff costs while incurring minimal in royalty admin fees, payment for usage of others’ algo IP, and subscription for pattern monitoring services. Existing vested interests, other encumbrances, and new entrants can all flourish under this recommended approach regardless of interactive or non-interactive subscription services they choose. 
  • Policy makers should welcome this approach because the only way to regulate the decentralized equity markets is via decentralized crowd intelligence.