What should the Tick Size Pilot be about?

Virtually all market participants agree that the notion of "One Size Fits ALL" that currently describes the US equity market structure is flawed. The implication is that large cap equities trade exceptionally well in the U.S., but smaller capitalization stocks have suffered (at least on a relative basis) in terms of available liquidity.

Ostensibly in response to this concern, at the direction of Congress, the SEC ordered all of the nation's stock exchanges to develop and jointly file an NMS pilot plan to evaluate changes to this policy. This "Tick Size" pilot is supposed to study the effects of wider tick sizes for less liquid stocks.

I believe that there is a binary choice that needs to be made regarding the pilot. Either the pilot should be designed to widen SPREADS, or it should be designed to better relate tick sizes to prevailing spreads without widening them necessarily. The latter option is a direct method of analyzing "one size fits all" and is clearly what I favor.

The stated goal of the Pilot is to "assist the Commission, market participants, and the public in studying and assessing the impact of increment conventions on the liquidity and trading of stocks of small capitalization companies." Despite the clear intent to analyze "one size fits all" as it relates to tick sizes, there seems to be some confusion regarding the key objectives of the pilot.Some have argued that we should study if wider spreads would either improve liquidity or, by increasing profits at market makers, increase analyst coverage by those firms. On this point, I completely agree with Larry Tabb in his recent post (http://tabbf orum.com/opinions/why-the- tick-pilot-has-little- chance-of-spurring- permanent-change) that the pilot will both fail to achieve this goal and be too short to measure that effect in any case.

Our view at CoreOne Technologies is that the best way to assess the impact of increment conventions on the liquidity and trading of stocks is to study the effect of varying tick sizes relative to the average displayed spread. The hypothesis is that tick sizes that are a larger percentage of the displayed spread will improve liquidity by providing increased incentives to quote. There are clearly defined metrics to test this including the change in the actual average spread, the effective spread for executed orders, the amount of liquidity displayed, and the actual volume traded relative to the market cap of the security. If effective spreads decrease and liquidity increases in the test group, then that would essentially prove the hypothesis true. In that case, we would have gained important evidence that tick size is a statistically significant factor for adjusting market structure, and how appropriate tick sizes would improve pricing and liquidity for smaller cap stocks.

Determining which goal the Pilot is looking to achieve has very clear implications for the design of the Pilot itself. If the goal is to simply widen spreads, then, despite objections from the industry, market maker profitability and analyst coverage data both would become necessary metrics to study. It would also mean that the current design of the pilot, which includes a significant number of issues that already trade with average spreads below 5 cents, is reasonable. This is all beside the point. The goal SHOULD be to test the hypothesis that relating tick size to the average quoted spread would improve market structure. To achieve this goal, the pilot universe should be redesigned and the metrics should be clarified.

CoreOne would propose a re-design of the pilot universe. The universe should be restricted to small cap securities where the average displayed spread in the previous quarter was greater than 5 cents. If there was scope to consider different tick sizes, we would suggest adding both a 2 cent tick size and relaxing the restriction to stocks where the average spread was greater than 2 cents, and potentially a 10 cent tick size for those issues whose average spreads are greater than 20 cents. This approach has two important benefits. First, it would allow the industry to measure the effect on liquidity directly without necessarily increasing trading costs to those investors (mostly retail) whose trading costs are directly related to the spread. Second, it would shift focus to easily defined metrics such as effective spread, percent of market capitalization traded, and average liquidity displayed instead of trying to assess market maker profitability and other hard to quantify statistics.

In addition to changing the pilot universe, CoreOne would further suggest a simplification in the data collection strategy. We believe that the most efficient way to measure the pilot is to leverage the existing infrastructure for rule 605. That infrastructure, which is already in place at most firms, measures effective, quoted and realized spreads, fill rates, price improvement and captures most important execution information. Firms that generate these statistics have access to all relevant liquidity information and have experience in generating statistical analysis defined by both order size and stock groups. In addition, this analysis is published with sufficient delay to alleviate concerns of information leakage that many participants are wary of. Thus, we propose that the SEC should, for the Pilot universe, remove all exemptions from 605 reporting that derive from parent order instructions such as "not held" or parent order size (Restrictions for orders that have special instructions like All Or None, which constrain execution would not be effected in this proposal). This would create a relatively complete dataset of all market, marketable limit, and non-marketable limit orders sent to exchanges, ATSs, broker dealer internalization systems and market makers. Since institutional, proprietary trading, and algorithmic orders would be added to the current 605 universe that is focused on retail orders, it would allow independent analysis of the key trading characteristics that would determine the truth of our hypothesis. If the rule for the pilot universe also removed exemptions for all non-displayed, pegged and liquidity seeking orders sent to dark pools, the result would be a relatively complete dataset for analysis. This approach would be far less expensive to implement than the currently contemplated framework and would be capable of generating the key metrics required.

The bottom line is that the pilot should focus explicitly on improving market structure for smaller companies. It should not try to widen spreads for the purpose of placating those that yearn for the halcyon days of market-making.

S&P Global provides industry-leading data, software and technology platforms and managed services to tackle some of the most difficult challenges in financial markets. We help our customers better understand complicated markets, reduce risk, operate more efficiently and comply with financial regulation.

This article was published by S&P Global Market Intelligence and not by S&P Global Ratings, which is a separately managed division of S&P Global.