It seems the law of unintended consequences is hitting NasdaqOMX- right where it hurts in the volume basket.
How? By its own hand and its recently enacted price pilot program.
On February 2, Nasdaq initiated a pilot program for experimental pricing for 14 symbols: 7 Nasdaq -listed and 7 NYSE-listed. For this group, Nasdaq adopted a new 5/4 pricing model – replacing the existing 30/29. The stated goal of the pilot was to improve overall market quality and, in doing so, direct a larger portion of trading to their exchange away from both lit competitors and off-exchange platforms such as ATSs.
For the pilot program, ITG noted that Nasdaq picked very liquid stocks for the pilot, with the least liquid stock still trading over 7.5 million shares per day. For the NYSE securities, each traded over 10 million shares per day.
“This makes sense because the test would theoretically have more of an effect on liquid securities as there is often much more competition amongst the exchanges on the NBBO,” ITG wrote in a research note.
However, ITG found upon analysis that overall market quality impact appears minimal in spread sizes and top of book depth and that after 18 days of the pilot, results showed a “measurable” decrease in NASDAQ market share, both in quoted volume and traded volume. This comers despite ITG’s passive trading, which through its SLimit algorithm, saw increased trading at NASDAQ as a result of shorter queues compared with historic averages.
ITG explained that to analyze the effect of the change in rate, it compared the pre-pilot month of January 2015 with the results of February 2- February 25 (18 trading days) for the pilot stocks. The goal was to test the theory that the change should result in measurable improvement to price discovery, as well as whether or not Nasdaq benefits from the resulting effects. ITG focused only on quoted and traded volumes on lit exchanges, considering the following statistics: Average Spread, Average Depth of Top of Book – both consolidated and NASD-only and Nasdaq’s market share in shares traded.
“We hypothesized that the results would be insignificant, assuming that most market participants would not invest the resources to adjust routing behavior for such a small universe,” ITG wrote. “The results were surprising in that there were definite changes to the market structure and quality. This demonstrates how price sensitive some market participants have become.”
For the effect on market quality, ITG focused on average spread size and average depth at the top of book. The average spread before and after the change it found was that virtually identical for both the NYSE and Nasdaq-listed names.
“Interestingly, these changes in overall quoted volume were seen primarily among Nasdaq’s competitors. In January,” it added.
Nasdaq had been about 26% of the NBBO quoted size before the pilot program started, according to ITG. But that same number shrank to 20% in February. The same held true for NYSE stocks – Nasdaq’s market share of quoted size decreased substantially, from 15.1% to 10.4%. “It seems curious that the overall top of book would be largely unchanged with such a drastic decrease in NASDAQ’s market share.”
ITG opined that market share in traded volume is probably the category most important to Nasdaq, and here the firm saw a big decrease. If the symbols are weighted equally, ITG said Nasdaq market share dropped from 36.4% to 30.2% on the Nasdaq -listed stocks and from 20.3% to 17.1% on the NYSE-listed stocks.
On the whole, ITG cautioned that it is too early to discern whether this has had a positive or negative effect on the market from a market quality perspective. It noted that the effect was most pronounced in some symbols in the first few days of the pilot.
“It is possible that market participants were waiting to see some results before adjusting their routing strategies,” ITG noted. “. What does seem more certain is that the pilot succeeded in decreasing NASDAQ’s market share, which was surely not their intention. NASDAQ will have to market these changes to institutional brokers for it to pay off, as the HFT community seems to disfavor lower rebates in all but the deepest-queued names.”
Given these results, ITG said the pilot has merit and is worth continuing as the buyside has become much more interested in reducing or eliminating maker/taker fees. However, ITG theorized that the only market participants that appear to have adjusted their trading strategies the fastest are high-frequency traders.
“This has allowed firms like ITG to achieve higher queue positions in NASDAQ to the benefit of our clients,” ITG said.”The brokerage community, if interested in improving performance on behalf of their clients, should pay heed.”
ITG concluded that it welcomes the expansion of the Nasdaq pilot to more stocks to see the implications.
“Without some kind of rallying cry prompting more global adoption, the idea will probably fall by the wayside, as indicated by the decreased market share we have seen so far.”
