College of Humanities and Social Sciences

Three Essays on Naked Short Selling and Fails-to-Deliver

John W. Welborn

Major Professor: Thomas Stratmann, PhD, Department of Economics

Committee Members: Alex Tabarrok, Carlos D. Ramirez, Stephen E. Christophe

Carow Hall, Lecture Hall
June 07, 2013, 10:30 AM to 07:30 AM


This dissertation consists of four chapters that investigate the causes and consequences of fails-to-deliver (FTDs) in U.S. stock markets. In Chapter 1, I present a brief history of U.S. trade settlement institutions. In Chapter 2, I analyze the effects of eliminating a market making exception to timely close-out requirements on FTDs and stock borrow rates. In Chapter 3, I show that amendments to SEC short sale rules reduced common stock FTDs but led to higher FTDs in exchange-traded funds (ETFs). Further, positive changes to ETF FTDs Granger-cause higher market index volatility. In Chapter 4, I find that high FTD stocks experience abnormal negative returns, and thus high FTDs indicate a nonbinding short sale constraint.

In Chapter 2, I investigate the consequences of eliminating the Options Market Maker Exception to SEC Regulation SHO (the “Exception”). Until 2008, options market makers that engaged in bona fide market making were exempt from locate and certain close-out requirements for short sales. The Exception applied only to short sales that qualified as bona fide hedges of options positions that were established before a stock went on the Regulation SHO Threshold List. I test the hypothesis that eliminating the Exception reduced the incentive to naked short sell stocks through the options market. I compare data from the second and fourth quarters of 2008. Consistent with my predictions, I find that eliminating the Exception led to fewer FTDs and higher stock borrow rates for optionable stocks as compared to non-optionable stocks. Further, removing the Exception reduced optionable stock FTDs when the price of borrowing stock was high. Finally, options market trading volume declined after the Exception was eliminated.

In Chapter 3, I investigate the determinants of ETF FTDs. ETF trading volumes have increased over the last decade, and so have unsettled ETF trades at the clearing corporation. ETF FTDs are large and persistent despite SEC rules that require timely close-out. I document positive relationships between ETF FTDs and short sale volume, stock borrow costs, put option open interest, and quarterly index options expiration (“triple witching”) dates. These findings are consistent with the hypothesis that market makers fail to deliver to avoid borrowing costs associated with short sales. I also document a positive relationship between short sale demand and changes to ETF shares outstanding. I then find that positive changes in aggregate ETF FTDs Granger-cause higher market index volatility. This is because market makers are required to buy or borrow stock to close-out ETF FTD positions by trade date plus six days (“T+6”).

In Chapter 4, I analyze the relationship between high FTDs and stock returns. The academic short sale literature views FTDs as evidence of binding stock lending constraints, and stocks with FTDs may be overpriced because short interest is below equilibrium levels. Conversely, high short interest stocks with nonbinding short sale constraints experience abnormal negative returns. This is because informed short sellers are willing to pay extra to short. I find that high FTD stocks from the Russell 3000 Index experienced abnormal negative returns from 2004 through 2008. I obtain this result in both an event study and a portfolio returns analysis using Fama-French factors. Thus, high FTDs are evidence of a nonbinding short sale constraint that does not restrict informed short selling because high FTD stocks, similar to high short interest stocks, experience abnormal negative returns. While this research does not determine whether FTDs depress stock prices, it demonstrates that high FTD stocks are not overpriced. Additional support for this finding comes from the fact that short interest and FTDs are highly correlated.

Print Friendly and PDF