The SEC's Short-Sale Experiment and Substantive Short Selling: Evidence on Causal Channels and Experiment Design

During 2005-2007, the US Securities and Exchange Commission (SEC) ran a randomized experiment, in which it removed short sale restrictions for some pilot (treated) firms and left these restrictions in place for others. Early studies of the experiment found limited direct effects on trading markets and no impact on share prices. However, the SEC experiment has recently been exploited by many finance and accounting scholars, who report evidence that removing short sale restrictions had a wide range of indirect effects. We show that, when studying the effects of substantive short-selling, the SEC, in effect, conducted four separate subexperiments, which should often be examined separately. We then develop theoretical reasons and evidence that the causal channels presumed by the indirect-effects studies do not exist, for all sample firms as a whole or for any of the four experiments. We also re-examine the evidence for effects on short interest and share prices for smaller firms, offered by Grullon, Michenaud and Weston (2015), and find no support for their results. We also explain the need, when studying indirect effects of the experiment, to account for nonrandom choices by short sellers on which firms to “target”, and the nonrandom responses of managers and other market participants to removal of short sale restrictions.