Sarbanes–Oxley Act and Patterns in Stock Returns around Executive Stock Option Exercise Disclosures (Digest Summary)

Regulatory changes in 2002 made the disclosure of exercising stock options by corporate insiders to the SEC within two days a requirement. Early exercise of options typically involves a loss of option value (“forgone option value”) to the holder because the option’s fair value tends to be greater than its intrinsic value (the stock price less the exercise price). Reasons for early exercise of executive options can be attributed to the need for portfolio diversification and liquidity, but a more significant reason is inside knowledge that the company’s prospects are deteriorating and a decline in the stock price is likely.

The authors use forgone option value to identify option exercise and subsequent trading that is most likely driven by the possession of negative private information. This information is shown to generate small, but statistically significant, negative returns since the 2002 Sarbanes–Oxley reforms compared with no significant response prior to the regulatory changes. Trading volumes tend to be higher in the post-reform period, while price drift over the subsequent three-month period tends to be lower than it was during the pre-reform period. Larger price movements are also associated with increased media coverage.

Previous researchers have considered the relationship between executive stock option exercises and subsequent investment returns; however, the changes to the option exercise disclosure regime make it possible to assess investor reaction to early exercising of options by corporate insiders. By using data from after the Sarbanes–Oxley Act and SEC reforms, the authors are able to assess the scale of investor reaction— in terms of both price reaction and trading volume—to the timelier disclosure information provided.

Although trading by executives can provide useful signaling information, it can also be information-neutral to the wider investment community. The classification of option exercises by forgone option value allows the authors to attempt to segregate trades driven by information-neutral motivations—for example, liquidity and portfolio diversification—from those potentially driven by the possession of negative insider information. To an extent, this classification relies on rational behavior on the part of the executives who are exercising their options.

This information is particularly relevant for equity and high-yield credit managers because the correct interpretation of any signaling provided by company executives can be invaluable in determining future returns.

Stock option exercise data are from the Thomson Financial database from 1997 (when the data are first available) to 2011. This window includes data prior to August 2002, when the SEC changed the disclosure rules to require companies to report the exercise of company options within two days.

The forgone option value resulting from early exercise is calculated using the standard Black–Scholes option valuation formula, despite corporate options not strictly satisfying all of the underlying assumptions of the pricing model.

Abnormal stock returns are calculated as the difference between the stock return (including dividends) and the expected return, which is calculated based on its exposure to four fundamental investment factors using the standard Carhart four-factor model approach.

Abnormal buy-and-hold returns are then calculated for a four-day period commencing at the SEC filing date. The data are then further split into three groups by the level of forgone option value. A final test compares differences by company size. The authors evaluate pooled results and test a separate regression approach with the abnormal four-day returns regressed against explanatory variables, including size, forgone option value (as a percentage of total assets), net proceeds of the subsequent stock trades (again as a percentage of total assets), and reporting lag. Industry and year effects are also captured.

Volume effects are tested using a regression approach with explanatory variables, including general market volumes, day-of-week effects, proximity to holidays, and flags indicating earnings or dividend announcements.

Finally, to assess the impact of varying levels of media attention on the results, articles that include text relevant to option exposures are identified and events are classified accordingly. This is an interesting area for future study and could have been fleshed out in more detail using more sophisticated text-processing methodologies.

Transactions in a company’s shares by corporate insiders have always been of interest to investment managers as a potential indication of the company’s prospects, but many motivations can be behind insider transactions that are independent of the company’s future direction, including diversification and liquidity.

By estimating the value that is lost through early exercise, investors can assess the likelihood of the trade being driven by concerns about future prospects and reposition their portfolios accordingly. Making portfolio decisions based on forgone value, however, remains challenging because doing so relies on corporate insiders acting rationally with respect to management of their own personal finances. Numerous studies have suggested that is not always the case, even with potentially more financially savvy individuals.

Although the data used to estimate the universal forgone option value are universally available, determination of an appropriate discount to apply to the valuation of affected stocks remains challenging because of uncertainty surrounding the timing and magnitude of any potential upcoming bad news.

Finally, given the relative arbitrariness involved in setting cutoffs for determining high forgone option values, it would have been interesting to track the actual stock returns and/or changes in profitability of those companies identified as highly susceptible to negative news and comment on the appropriateness of using the bottom 30% as the cutoff.