Both the number of restatements and the number of companies reporting restatements are declining according to a new study. The number of restatements has been declining for three years now, and the number has declined materially since the figures peaked in 2006, both because of better controls and changing standards.



The study, by Audit Analytics, is not yet available online, but it has been widely reviewed, including in a March 4, 2010 article (here) and a March 1, 2010 article by Matt Kelly of Compliance Week (here).



As reflected in this article, the study shows that there were just 630 companies reporting 674 accounting restatements in 2009. There were 24% fewer restatements in 2009 compared to the prior year, when there were 923. The 2009 figures represent the lowest number of restatements since 2001 (when accounting scandals dominated the headlines).



The number of restatements has actually declined for three years in a row since they reached their peak in 2006, when 1,564 companies filed 1,796 restatements. In other works, the number of restatements in 2009 was 62 percent less than the number in 2006.



In addition to the declining number of restatements, accounting errors requiring a restatement are now being caught sooner. The average restatement in 2009 covers a period of 476 days, compared to 716 days in 2006.



Restatements also reduced earnings by smaller amounts. 2009 restatements on average reduced earnings by $4.6 million, compared to $7.2 million in 2008 and $23.5 million in 2006.



The article reports that the study’s authors attribute the decline to two factors: improved internal controls as a result of Section 404 of the Sarbanes Oxley Act, and a 2008 recommendation by the SEC’s Advisory Committee on Improvements to Financial Reporting that the SEC “relax its requirements on what types of errors should trigger restatements.”



One circumstance supporting the suggestion that SOX may be contributing to the reduced number of restatements is the fact that the majority of U.S.-based companies issuing 2009 restatements (374 out of 522) were “nonaccelerated filers,” meaning that Section 404’s requirements do not yet apply to them. Of course, there are, in fact, more nonaccelerated filers than accelerated filers in the first place, so the raw numbers alone may not tell the whole story. In addition, the smaller nonaccelerated filers simply may be more likely to have problems due to their small staffs and fewer tools.



On his Compliance Week blog, Kelly points out that the number of restatements by accelerated filers grew between 2002 and 2005, the year they had to comply with Section 404, but they have declined since that time. Kelly concludes that, despite all of the criticism of the provision, Section 404 may be working.



To those who say we had a crisis in 2008 notwithstanding Section 404, Kelly points out that the most recent crisis “has largely been a crisis of flawed assumptions and reckless risk management coming home to roost – not accounting fraud.” Kelly concludes that whatever financial reform Congress might conjure up in response to the current crisis, it is not time to “start rewriting Sarbanes-Oxley wholesale,” as “the law is working just fine.”



The suggestion that the declining number of restatements is due to SOX reforms brings to mind the long-standing question whether the changes in the number of securities class action filings are also attributable to improved company behavior as a result of SOX.



However, though the number of restatements has declined steadily, the number of lawsuits has fluctuated from year to year. Indeed, the most recent year with the highest numbers of restatements, 2006, when there were almost three times as many restatements as in 2009, there were fewer class action lawsuit filings (116) than in any year since 1996, and certainly significantly fewer filings than in 2009, when there were (depending on whose count you are using) at least 178 filings.



So there may well be fewer restatements as a result of Sarbanes Oxley, but that alone does not explain what has been happening with fluctuating securities class action lawsuit filings. Changed corporate behavior as a result of Sarbanes Oxley, even if it has occurred, is not a sufficient explanation for lawsuit filing levels. There may simply be too many other areas of corporate activity, beyond those addressed in Sarbanes Oxley, that continue to attract the unwanted attention of the plaintiff’ class action securities lawsuits.



The bottom line seems to be that as good as the news is that the number of restatements is declining, that does not necessarily mean as a general matter that companies are necessarily less likely to be sued.