U.S. Securities and Exchange Commission 100 F Street, NE Washington, DC 20549-1090
Re: Publication of Certain Daily and Monthly Short Sale Data on the FINRA Web Site; File Number SR-FINRA-2009-064
Dear Ms. Murphy:
The Commission recently published for comment a "Notice of Filing and Order Granting Accelerated Approval of Proposed Rule Change" relating to the publication of certain daily and monthly short sale data on the FINRA web site. According to the filing, FINRA is not proposing any changes to the By-Laws, Schedules to the By-Laws, or Rules of FINRA.
The purpose of this comment letter is to raise a concern about the data FINRA proposes to publish. The filing notes that the "SEC staff stated that it believes that the publication of this data, and the resulting increased market transparency, may help bolster investor confidence and thereby help promote capital formation."
This is a laudatory goal. However, it is important to note several factors about the data that may impact market transparency, and therefore investor confidence, relating to Commission and FINRA reporting requirements. These factors are in addition to the possible data imperfections pointed out by FINRA in the filing. Presumably the intent of daily and monthly short sale information is to make transparent the volume of stock actually sold short, thus impacting the amount of short interest in the security, and reflective of the amount of stock that must be borrowed. However, with respect to broker-dealer proprietary sales, especially when acting in the capacity of market makers and block positioners, such trade volume information may not always depict accurately the quantity of stock sold short. It may tend to over-count such volume, and therefore, if published, unnecessarily impact investor confidence in an unforeseen way. To be clear, I am not suggesting that no data be published. Rather, the Commission and FINRA should consider reviewing and addressing short sale recording and reporting requirements so they are more meaningful.
One of the primary functions of broker-dealers is to act as intermediaries for investors that are buying or selling stock. Often, to carry out that function, broker-dealers will handle such investor orders on a riskless principal basis. A riskless sale is one in which a broker-dealer, after having received an order to sell a security, sells the security as principal, at the same price, to satisfy that order. Regulations require broker-dealers to mark their proprietary riskless sell order as short if they don't own the security, even if the customer order to sell the security is long. Since broker-dealers generally don’t maintain a position, a significant number of such riskless sales are reported as short, even though the customer is selling long, and the broker-dealer intends to and will buy the shares from the long selling customer immediately after the proprietary riskless short sale takes place. Typically, the broker-dealer's position is short for considerably less than one second.
I don't know how much of these short sales can be attributed to long sellers where broker-dealers are effecting the transactions on a riskless basis. But FINRA knows, or at least could make a good estimate using OATS information. An OATS execution report is required, among other things, for every customer order filled on a riskless basis and related OATS records indicate whether the investor order was for a long or short sale. It might be sensible to conduct such a review prior to publishing this data.
In addition, the SEC recently updated their guidance concerning how sell orders are to be marked, indicating that pending sell orders, but not pending buy orders, be taken into account when determining whether a sale is long or short. This guidance does not take into account the probability that the orders will be executed, or the firm's position at the time of execution. As a result of such guidance proprietary and market maker sell orders and trades will be marked short when the firm is actually long, not taking into account pending sell orders. At the end of the day, the firm may well end up flat, meaning that they have no position in the instrument.
Accordingly, the marking of sell orders as short by market makers, block positioners, and broker-dealers effecting riskless principal transactions may not always accurately reflect what has transpired, and publishing data derived from these transactions may not increase market transparency and bolster investor confidence. FINRA appears to acknowledge these constraints by noting in the filing that
"information relating to market maker [or supplemental liquidity provider] status is not currently included in the trade report submission; thus, FINRA currently is unable to separately identify the trades of equity market makers [and supplemental liquidity providers] in the monthly short sale transaction file. Therefore, the "short type" field will include a value of "S" in all cases."
FINRA knows, or can determine, whether a firm is a registered market maker in specific instruments. However, determining whether a firm is a block positioner may not be practical. A more straight-forward way of recording and reporting sell orders and transactions by market makers, supplemental liquidity providers, block positioners, and other broker-dealers effecting riskless principal transactions could be considered by the Commission and FINRA.
Since the reporting of short sales by these market participants is problematic and not useful in terms of market transparency, and since, importantly, these market participants are generally afforded exemptions from having to borrow stock when effecting short sales, all such sell orders and trades could be marked with a specific designation other than that of a long or short sale, such as a sale made or order entered in connection to market making, riskless trading, or acting as a block positioner. Then, the daily and monthly short sale information published by FINRA would in a more meaningful way reflect true short selling activity taking place.
News 
Market Data 
Discover 
