One of the provisions of the JOBS Act of 2012 strongly suggested that the Securities and Exchange Commission experiment with the tick sizes of securities. There is an impressive case to be made that wider tick sizes support an infrastructure that has various positive effects that have been squelched by the 1990s move to doing everything in pennies or smaller increments still.
Accordingly, in February 2014 the House of Representatives passed a bill more specifically targeted on this point than the broad gauge JOBS Act had been. The House passed the Small Cap Liquidity Reform Act, the brainchild of Reps Sean Duffy(R-WI) and John Carney (D-DEL). The Senate never passed this bill, but even House passage alone seems to have spurred the SEC into getting back to work on the subject.
In June, the SEC announced a pilot tick-size program for U.S. equity markets.
But this has created a lot of tricky an issue over details of the pilot, or actually part of the pilot, for the program is supposed to entail three tracks. One of the three pilot programs is to have a “trade at” rule.
Ah There’s the Rub
A trade-at rule ensures that aggressive liquidity providers are the first to trade, because it prohibits brokers, internalizers, or dark pools from trading before a posted quote unless there is a significant price improvement. Under such a rule, (as Larry Tabb recently put it in a paper on the issue at TabbForum) a broker will have little flexibility in the execution of orders. If a broker is unable to match a client’s order with significant price improvement, it will be “forced to route the order to the market with the best protected displayed price.”
Exchanges will have an analogous loss of flexibility. They will be prevented from leveraging undisplayed liquidity.
So: what is this “significant” price improvement by means of which the gate is kept? It is a full minimum price variation (MPV) or half the MPV in a one-MPV market. So for a nickel tick pilot, the smallest significant price improvement is half of that, or $0.025.
Here is an example, drawn from a comment letter submitted to the SEC by the Security Traders Association. Suppose ABC is a security within Test Group Three, the Trade-At part of the program. The National best bid and Offer for ABC is $20. X $20.10. A particular trading center (TC1) displays a 100 share protected bid at $20. Another (TC2) displays likewise. But TC2 also has 300 shares hidden at $20. There are no other protected bids.
Given those facts as background, suppose an incoming order arrives at TC2, a sell order for 400 shares. That is an amount (including both protected and hidden) that matches the amount bid for. But TC2 would not be able to satisfy the entire 400 shares under the trade-at rule. It would execute the first 100 against the protected bid. Thereafter it must respect the protected bid at TC1 for $20.
Just Test the Simple Issue Already
This is the sort of situation that commenters see as an anomalous consequence of the proposed rule. At a minimum, critics contend, the Trade-At rule for this group introduces complexities that direct from the fairly simple issue that the pilot is supposed to be testing.
A related issue, also raised by Tabb, is that retail brokerage firms within the pilot will have to attest “that their flow is manually generated and not automated or algorithmic.” This, Tabb says, “is harder to determine than one might think.”
For example, a market maker may employ a trading model according to which a stock is purchased once it crosses its 10, or 20, or 200 day moving average. This model may work in a number of ways. When the price crosses the threshold of the model, the crossing may merely set off an alert, or it may automatically generate an order, or it may do something between those two possibilities.
Tabb cautions that the Trade-at rule will force brokers to “differentiate these scenarios,” and it isn’t clear what is gained by the expense involved in doing so.
For these and related reasons, he believes that the Trade-At provision “will benefit few, and harm many.”