Silence is golden -- so why do firms disclose?

Firms can avoid damage from accounting restatements simply by disclosing the restatement less prominently, according to a study by Swanson, Tse & Maynalda, Stealth Disclosure of Accounting Irregularities: Is Silence Golden? Here's the abstract:

Managers are allowed considerable discretion over how they announce an accounting restatement. Some firms issue a press release that discloses the restatement in the headline (high prominence); others provide a press release with a headline on a different subject (for example, earnings news) but discuss the misstatement in the press release (medium prominence); and most of the remaining firms simply change the comparative-period amounts reported in an earnings release, with no direct mention of the restatement (low prominence). Mean three-day returns are negative and differ substantially across these categories of disclosure prominence (-8.3%, -4.0%, and -1.5% for high, medium, and low prominence, respectively). We investigate whether this pattern is due to differences in the severity of the accounting irregularity or to the relative transparency of the three press release formats. Our tests support the latter explanation. We also find that companies providing less prominent press release disclosure of their restatement are less likely to be sued for securities fraud. In sum, both return and litigation tests indicate that ?silence can be golden.?

The study has several interesting implications.

First, it suggests that firms themselves are a more important source of information about disclosure problems than enterprising class action lawyers. In other words, the lawyers seem to be mainly reading the newspapers rather than ferreting out fraud.  This obviously bears on the social value of securities class actions.

Second, if firms can reduce the risk of legal action by staying quiet, the securities laws would seem to discourage rather than encourage disclosure.

Third, if firms are acting rationally, extralegal incentives such as reputation must be important in encouraging disclosure -- perhaps as important as the securities laws.

These implications fit nicely with my theory of the relative roles of markets and securities laws in disciplining fraud.