Congressional Market Structure

While a Pilot Program focused on the most liquid stocks is not a bad idea, the notion of price controls as a singular solution, is.

Many of the most liquid stocks suffer from a problem insofar as they naturally trade at a narrower spread than the one cent increment mandated for all stocks. That said, the Congressional bill would result in an increase to the actual spread in those stocks, by eliminating the markets adaptation of "inverted" venues.

To illustrate this in SIRI:

Market spread is usually 0.68 cents due to the fact that the BATS Y exchange PAYS liquidity takers 0.16 cents to remove liquidity. This bill, were SIRI included, would increase its displayed spread in the market by 47%! Yes - that is correct, this proposal would increase the net spread paid by traders in the markets by almost 50% in the most liquid securities!

Put simply, this bill, if it exclusively focuses on "maker taker," will serve to eliminate a significant market adaptation to the current "one size fits all" policy in the US equity market. The one cent spread for all equities is a main reason that there is such a wide range of access fee /rebate combinations across stock exchanges and trading systems. Exchanges do this in order to attract order flow across a wide range of stocks. Google, at over $500, trades very differently than Apple at $120 and both trade differently than SIRI, which trades at around $3. It should not be surprising, therefore, that the market share of different market centers and exchanges vary a great deal from stock to stock. Markets that provide rebates to liquidity takers tend to attract more order flow in active securities, where the one cent minimum spread is artificially wide (such as in the case of SIRI). Markets that provide large rebates to liquidity providers tend to attract more order flow in stocks with "natural" spreads of one cent or more.

In addition to this particular bill, there have also been calls to cap access fees. This type of proposal is also problematic for many of the same reasons. If it is determined that a pilot program for active stocks is indeed warranted, then the most sensible pilot is to test sub penny pricing, rather than fee caps or banning rebates. The current fee cap of 0.3 cents is equal to 30% of the 1 cent tick-size. IF tick size is reduced to 0.2 cents (from 1), keeping the fee cap at the same ratio would mean the fees would be capped at 0.06 cents, which is almost the same as the proposal to cap fees at 0.05 cents. Thus, allowing sub-penny pricing in these stocks would achieve materially similar cuts in access fees as a hard cap. This approach, however, would decrease spreads for the most liquid securities and likely reduce trading costs for investors. It would also have the benefit of NOT restricting competition and innovation, which have been the hallmark of our markets evolution.

It is also important, when considering the idea that we should ban rebates, that one of the main beneficiaries of rebates are retail investors. Whether their clients benefit indirectly through enhanced services or directly via lower commissions, retail brokers collect rebates for their order flow. Thus, it is a virtual certainty that banning maker-taker would lead to increased costs to retail investors.

Another concern is that the banning of maker taker could actually gain enough support to lead to a broader application. A broader ban would have the effect of reducing market making incentives for small to mid-cap securities. As I pointed out in the Grand Bargain commentary, such an idea runs directly counter to the JOBSAct by reducing liquidity for small companies, thereby, creating even larger structural hurdles.

Smaller companies display a demonstrable decrease in liquidity, relative to large caps, due to tick size being too small, compared to the bid offer spread. (Comment /data). This change may seem small to the uninitiated, but the difference between maker and taker fees is more than half of the tick size. This means that a broad ban of maker taker essentially cuts the effective tick size for many stocks by 50%. Thus, doing so across the board could have far reaching impact and should certainly not be done by legislative fiat.

Based on this bills introduction, it would appear that lobbyists have convinced Congress to take money from retail investors in order to help "for profit" exchanges and large banks. Since this bill would result in increased spreads for extremely active stocks and prevent retail brokers from collecting rebates, it is hard to justify. At the same time, giving credibility to the "benefits" of eliminating rebates altogether could lead in the future towards higher trading costs for small to mid-sized companies, as the incentives for providing liquidity are reduced.

As I have stated before, there is a better answer, Transparency. Improved disclosures and execution analysis are the best way to resolve conflicts of interests and improve market quality. As the old saying goes: "Sunlight is the best disinfectant".

I agree with the BATSs letter to the SEC on this point and strongly urge Congress to let the SEC focus on improving disclosure and transparency. The alternative spelled out in this bill, while "just a pilot" is simply an example of anti-competitive price controls.

Given the track record of such legislation, one would hope we had learned that price control regimes end badly.

David Weisberger, Managing Director, Trading Services at Markit
Posted 3 June 2015

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.