Abstract:
The SEC allows market makers in U.S. equity markets to sell shares they do not hold in inventory in order to maintain a smooth and orderly market during periods of abnormally high volume. However, if the dealer is unable to locate and deliver the shares within the standard T+3 settlement window then a delivery failure, or 'fail' occurs. In 2005, the SEC introduced several new rules in response to complaints that equity dealers were abusing their market making powers by selectively selling shares they did not own and allowing delivery failures to persist for weeks, months, and in some cases, years. The practice has been dubbed 'naked short selling' and the new SEC rules fall within Regulation SHO, or Reg SHO. Each exchange maintains a 'Threshold Security List' (TSL) containing all the stocks on the exchange that meet Reg SHO criteria for delivery failure.
This study begins with a statistical summary of delivery failure in U.S. equity markets. Fails are disaggregated by exchange, month, firm size, duration and other fundamental and technical variables. Event studies highlight the relationship between price-volume spikes and deliver failure. Statistical inference attempts to assess the degree of success of Reg SHO and address the charge that dealers target fundamentally weak companies.
Preliminary results indicate that the average duration of an open fail is 16 trading days beyond the T+3 settlement date. The number of securities on the TSL on any given day is small compared to the thousands securities on the major U.S. stock exchanges. On average, close to 2% of listed companies are on the TSL of one of the 5 major U.S. exchanges, NYSE, Nasdaq, Amex, OTCBB and Pink Sheets. However, a significant fraction of companies meets the TSL criteria at some point during the year. For example, during 2006 the number of unique companies appearing on a TSL ranged from a low of 9% of Nasdaq listed securities to a high of 15% of Amex listed securities, including mutual funds, a major component of Amex TSL securities. The number of firms on a TSL and their mean duration is relatively stable over the sample period, an indication that Reg SHO has had limited success in reducing delivery failure in U.S. equity markets. Event studies indicate that large price-volume spikes are a key factor in the creation of open fails. On average, technical factors appear to explain more variation in delivery failure than fundamental factors.