This morning, the CHX published a response to the comments and letters objecting to their LEAD proposal to introduce a speedbump. You may recall that the CHX proposed to delay all orders and cancel requests for everyone other than their qualified market makers, thus providing those market makers an unnatural advantage. In essence, it gives them a leg up in the time/space continuum, as Doc Emmitt Brown might say. That’s why it inspired my “Back to the Future” blog in the first place.
Sadly, today’s CHX sequel didn’t do much to allay my concerns. In fact, I found the logic so tortured, that I was reminded of the plotline for the Back to the Future sequel.
As I was reading the CHX analysis that essentially said:
“PLEASE give our market makers substantial advantages while protecting our quotes, and I promise that they will make tighter spreads for more volume than their competitors”
I kept thinking:
“Just tell our market makers what teams will win and lose in every sporting event before they happen and they’ll create an amazing boomtown for everyone.”
But, like the mess Biff made of Hill Valley, it won’t work.
CHX’s own data shows that, even before their market makers detected what they term “latency arbitrage” that their market share for quoting was much higher than for trading. Despite averaging a quote market share on either the NBB or NBO of roughly 24% in SPY they only averaged a trading market share of 5.7%.
Thus, it is reasonable to assume that CHX market makers, who earn NMS quote revenue sharing based on their quoting activity, whether or not they trade, were more interested in quoting than trading. It is also interesting that, after these market makers detected “latency arbitrage” that, not only did their quoting frequency and size drop, but that their trading market share dropped much more. In fact, according to the CHX’s own data, their relative market share of trading compared to quoting dropped by roughly 66%. This strengthens the argument that they are indeed motivated to quote without trading.
Thus, I would argue that the historical behavior of CHX market makers makes it very unlikely that approval of LEAD would lead to tightening of the aggregate market spread. Of course, the CHX themselves already said as much when they noted “While the Exchange did not observe any discernable change on the NBBO spread in SPY during the After Period” in their own comment letter. Thus, it is reasonable to conclude that the advantages claimed for approving LEAD would not materialize.
There is also a problem with their notion of Latency Arbitrage; Whenever the market is moving there will always be a “last quote standing”. It is not necessarily an arbitrage to trade with that last quote, but it is certainly not advantageous to a market maker to be the last quote providing liquidity. Such quotes do suffer adverse selection as the spread tends to move against such quotes after they are executed. As a result, I certainly agree with the CHX that being a market maker on an exchange that is consistently at the end of firms routing tables, is a bad place to trade. Considering the location of the CHX matching engine in Chicago introduces significant latency, and routers prioritize speed when selecting the order of venues to route to, it is an unenviable position indeed.
That said, what CHX is asking, is to allow their LEAD MMs to be able to avoid, in all cases, being the last quote. What they are NOT saying is that the second to last quote becomes last, if the LEAD quote fades. Since every market maker would, of course, want this advantage, if approved, EVERY exchange could do this, turning the whole NBBO into a farce as market makers would all choose to have the option to fade their quotes.
However, if the SEC eliminated the OPR AND excluded the LEAD quotes from the SIP NBBO to avoid the risk of publishing fake spreads, then it would be reasonable to revisit the CHX proposal. Without forcing the rest of the market to route orders to the CHX and use potentially phantom quotes as a benchmark for trading, then the LEAD proposal would not damage the integrity of the market.
If, after approval in such a scenario, the LEAD MMs offer more size, tighter spreads, and those quotes did not fade in practice, then the best execution obligations of brokers will turn the CHX into a success story. That is the path that NEO has taken in Canada and is a better model. But, if the SEC approves this proposal with the OPR intact, it will be a setback for the years of progress made in making trading more efficient.
Time travel may make for nice, escapist fantasy, but I don’t think any of us want to take our trading back to the period before Regulation NMS leveled the playing field. The SEC should deny this request.
 SECURITIES AND EXCHANGE COMMISSION (Release No. 34-80041; File No. SR-CHX-2017-04) February 14, 2017, CHX filing, pages 32-35 (www.sec.gov/rules/sro/chx/2017/34-80041.pdf)
 Ibid, page 34