European Fragmentation = “Back to the Future”
There have been a series of articles on the likely Fragmentation of the European equities market such as coverage from Markets Media and TabbForum which explain that 20 or so new Systematic Internalizer (SI) venues will be providing liquidity in 2018. These venues will be able to quote differently per client (or client type) and can stream these quotes directly to clients or systems such as SORs or Multilateral trading facilities. This new regime features capital commitment as its centerpiece, with the ability of principal trading firms to segment clients based on their perceived dealing risk. While this will certainly be a major increase in fragmentation, it is not really new.
The development of SIs reminds me, in many respects, of the old SEAQ International system that operated in London in the early 1990s. In those days, most of the European markets operated Central Limit Order Books (CLOBs) without direct competition in their home markets, but the London dealers competed with them by offering principal liquidity. The “system” for doing so was SEAQ International, which allowed dealers to post quotes for a standard order size. The system, however, did not have automated execution, nor did it offer any variations in quotes like the modern SI regime. It did, however, allow for segmentation as the dealers, upon receiving a call from a buy side client or a broker representing a client, could tailor their principal bid or offer to that client. Despite having wider quotes than the local markets and not much more displayed size, SEAQ International had very significant market share at the time, as European buy sides enjoyed the immediacy and certainty of the principal trading it facilitated.
As a result, the development of Sis does have a “silver lining” to institutions, who feel like they are not treated well by a fully order-driven market. The most likely evolution will be that Sis will display quotes for individual client types, and perhaps also on the specific client strategy. While this will undoubtedly require a large upgrade to the Smart Order Routers offered by agency brokers, the end state might be the type of customization that smart buy sides can take advantage of.
MiFID and Reg SCI testing requirements will likely provide regulators an open checkbook
Articles such as yesterday’s FIX Global piece which makes the point that testing algorithms and systems is already part of most firms software process, are somewhat in denial. All capital markets would breathe easier if this was true, but it is extremely unlikely due to the lack of realistic simulators of exchange and MTF functionality. The reality is that most algorithm errors occur as the result of edge conditions, where a particular exchange behavior or market data condition results in either order(s) exceeding the available liquidity or algorithms that go into a loop when it does not recognize when their orders were executed. To prevent these situations, realistic simulation environments, complete with software that mimics the changes to exchange order books is required. Sadly, apart from a startup company I have been talking to, I know of no commercial solutions that offer such functionality in the equity markets. As a result, when firms inevitably have errors that impact the market, the regulators will be able to cite the MiFID or Regulation SCI rules and fine those firms to “rub salt in the wounds” of the trading losses such errors inevitably create.