Impact of the Evolving U.S. Equity Market Trends on Trading Model Inputs : Observations and Implications on Float and Volume Adjustments

In securities fraud class action litigations, trading models are widely used to determine aggregate damages for the purpose of negotiating settlements. To estimate damages, trading models require assumptions on the number of shares available for trade (i.e., the float) and the number of potentially damaged shares (i.e., the volume) of the at-issue security during the class period. However, a few of the key assumptions used in trading models are based on an institutional framework for equity trading that is clearly archaic.In this paper, we examine changes in the institutional framework of U.S. equity trading since the early 2000s and address their implications on the float and volume inputs of the trading models. These developments include (i) the shift towards passive investing, and (ii) the adoption of SEC Regulation National Market System (“Reg NMS”) and the subsequent rise of high-frequency trading and dark trading, which affect the number of traded shares that are eligible for damages claims. We find that the prevailing methodology can result in an unrealistically high level of float that implies nonexistent shares available for trading for heavily shorted stocks. An alternative approach that relies on the investment style and fund turnover rate of each institutional investor may improve the accuracy of float estimate. Furthermore, the prevailing methodology for calculating trading volume during the class period has deviated from actual market structure for U.S. equities. To better reflect the change in market structure for equities trading, we suggest adjusting trading volume based on the venues on which stocks are traded, as opposed to the exchanges on which stocks are listed. We expect that appropriate adjustments for float and volume would likely result in more accurate estimates of aggregate damages