A Closer Look at Financial Statement Restatements
Analyzing the Reasons Behind the Trend

By Lynn E. Turner and Thomas R. Weirich

DECEMBER 2006 - Financial restatements reached new heights in 2005, and the first nine months of 2006 restatements are ahead of last year. Just when the number of erroneous financial reports by publicly traded companies seemed to have peaked, they have continued to climb. The results show why investors can’t afford a return to pre-Enron securities regulation.

This article analyzes financial statement restatements during 2005 and the first nine months of 2006 obtained from a comprehensive analysis of nearly 25,000 company filings with the SEC. The article highlights the main causes of restatements, comments on restatements and securities litigation, and looks at the restatement rates of particular auditing firms.

The authors’ research focused on restatements filed to correct accounting errors and, therefore, did not include restatements for changes in accounting principles (e.g., GAAP-to-GAAP changes, changes in estimates, or mandated adoptions of new accounting pronouncements). It also excluded restatements filed to add new discussion, to make minor wording changes, or to correct typographical errors. Because SFAS 154, Accounting Changes and Error Corrections, took effect for companies with fiscal years beginning after December 15, 2005, this study relied on the definition of an accounting error in Accounting Principles Board (APB) Opinion 20, Accounting Changes, where paragraph 13 states: “Errors in financial statements result from mathematical mistakes, mistakes in the application of accounting principles, or oversight or misuses of facts that existed at the time the financial statements were prepared. A change from an accounting principle that is not generally accepted to one that is generally accepted is a correction of an error.”

When restatements were identified, they were counted based on the underlying error or set of errors that drove the restatements, not the number of filings. Therefore, multiple filings that a company made to correct the same underlying error were counted as a single restatement.

Number of Restatements Reaches New Heights

Exhibit 1 presents a restatements scorecard for 2005. Companies with U.S.-listed securities filed 1,295 financial restatements, nearly double the previous year’s mark. This represents about one restatement for every 12 public companies (up from one for every 23 in 2004). Of these restatements, 100 were by foreign companies. The burden of these errors falls on investors, who rely on management to get the numbers right the first time.

These restatements aren’t just about revising subjective judgments or complying with esoteric, complex accounting pronouncements. In hundreds of instances, they stem from basic misapplications of simple rules or critical breakdowns in corporate controls and competencies. On the plus side, thanks to renewed urgency by auditors and their regulators, investors at least are getting better accounting on the next go-around. For that, investors can thank the renewed public-policy focus on the need to maintain strong internal controls.

The volume of restatements in 2005 likely would have been even higher had U.S. regulators not agreed to extend Sarbanes-Oxley Act (SOX) section 404 compliance deadlines into 2007 for smaller companies with stock market values of less than $75 million. According to the authors’ findings, the smallest companies are where strong internal controls arguably are needed most, because they are where the risk of restatement is the highest. Furthermore, companies audited by the smallest accounting firms restated six times more often than other companies—a trend that begs for more frequent inspections of these firms by the Public Company Accounting Oversight Board (PCAOB).

Finally, the authors also examined 2005’s surge in “stealth restatements”— companies that corrected material errors but with minimum publicity, by quietly including “fixes” in their quarterly and annual reports, rather than using the 8-K disclosure form that the SEC created expressly for announcing restatements. This should trouble any reasonable investor and demands the immediate attention of regulators. Exhibit 2 presents key findings from the study.

In 2005, 1,195 U.S. public companies filed financial restatements to correct errors, as defined under generally accepted accounting principles (GAAP), which increased 95% from year-earlier results. By comparison, from 1997 to 2004, the number of restatements grew at a 27% compounded annual rate. Exhibit 3 presents a historical view of the constant increase in restatements by U.S. public companies from 1997 to 2005. As of September 30, 2006, 1,042 U.S. public companies had restated, and 88 foreign private issuers, for a total of 1,130.

The restatements that concern investors most are the ones that cause a company’s sales, income, liquidity, or financial position to look dramatically different from the company’s previous reports. Not all restatements do this; however, it is better for a company to get its numbers right than to leave them wrong. While many corporate executives may have been slow to realize this, it has become more difficult for them to avoid it. And while investors may appreciate candor, the frequency with which companies have been incorrectly preparing their financial reports is no less disturbing. On a per-capita basis, restatements also climbed sharply in 2005, when the number of restatements, as a percentage of all U.S. public companies, was 8.5%, up from 4.5% in 2004.

From 1997 to 2005, U.S. public companies filed 3,642 restatements to correct accounting errors. That’s about 30% of all U.S. public companies over the past nine years. (The 30% figure is an estimate by Glass Lewis & Co. based on 3,642 restatements over a nine-year period and a weighted average of 11,333 U.S. public companies. Some companies restated multiple times. From 2003 to 2005, 320 restatements were by companies making at least a second restatement. Comparable data for 1997 to 2002 were not available.) When so many companies produce inaccurate financial statements, the quality of information that investors rely upon to make capital-allocation decisions is seriously called into question.

Financial reporting errors are most likely to be unearthed when companies’ independent accounting firms conduct their year-end audits. That helps explain why in 2005 March and April saw more restatements than in any other months. Of the 385 restatements filed in March and April 2005, 152 centered on faulty lease-accounting practices, primarily at restaurant companies and retailers. During these two months, 231 restatements were filed by companies that disclosed problems with their internal controls.

Exhibit 4 presents restatements by year, with year-to-date numbers as of September 30, for 2003 through 2006. Restatements for the first three quarters of 2006 are larger than the first three quarters of 2005. This is most surprising in light of the hundreds of leasing restatements during the same time period of 2005. If the trend continues, there could be as many as 1,500 restatements by year end 2006.

Restatements and SOX Section 404

Companies’ ability to produce accurate financial statements depends on the soundness of their internal controls over financial reporting. SOX section 404 requires that companies hire independent auditors to test the effectiveness of their internal controls and report their findings to investors. The first wave of these reports came in early 2005 when companies with calendar fiscal years filed their 2004 annual reports.

Executives’ complaints about these reports notwithstanding, requiring companies to have sound internal controls is nothing new. The Foreign Corrupt Practices Act of 1977 mandated that all public companies “devise and maintain a system of internal accounting controls sufficient to provide assurances that transactions are recorded as necessary to permit preparation of financial statements in conformity with generally accepted accounting principles” [U.S. Code Title 15, Commerce and Trade, Chapter 2B-1, Securities Investor Protection, section 78m(b)(2)].

Furthermore, since August 2002, SOX section 302, “Corporate Responsibility for Financial Reports,” requires top officers of all public companies to evaluate the effectiveness of their companies’ internal control systems and to include their conclusions in annual and quarterly reports. For approximately the first two years, the vast majority of corporate executives certified that their controls were free of any material weaknesses. As we now know, many of these companies’ controls did have major weaknesses. Not until independent accounting firms began to perform their own internal control audits did investors learn the extent of these problems. Often, the problems they discovered were serious enough to cause major errors on the companies’ financial statements, leaving management with no choice but to restate their previously issued financial reports.

The recent rise of restatements is the inevitable by-product of the financial housecleaning that was necessary to restore investor confidence in the wake of Enron, WorldCom, Tyco, Adelphia, and other corporate scandals. After more than 1,800 restatements during the past two years, the good news is that the major housecleaning may be nearly finished, although it may be years before annual restatement totals return to pre-2000 levels.

To some degree, the rise in restatements also may signal a culture shift among corporate managers, directors, and auditors. It has been more than eight years since then– SEC Chairman Arthur Levitt delivered his landmark “Numbers Game” speech, where he warned that the proliferation of earnings-management gimmicks had eroded the quality of corporate earnings and, thus, the quality of financial reporting. “Management may be giving way to manipulation; integrity may be losing out to illusion,” he said, calling upon corporate managers and Wall Street analysts “to embrace nothing less than a cultural change.” Today it appears at least some of the financial community has begun making that change, albeit belatedly.

Implications for smaller companies. Now that the major stock-market indexes have rebounded, the reform pendulum has begun to swing the other way. Over the past year or so, there have been a number of proposals and actions to scale back SOX requirements. In September 2005, the SEC for the third time postponed the deadline for non-accelerated filers (smaller companies) to comply with section 404, pushing it into mid-2007. In most cases, shareholders of these companies won’t receive an independent auditor’s report on the effectiveness of internal controls until early 2008—nearly six years after Congress passed SOX.

From the beginning, section 404 has drawn fierce criticism from a relatively small, but persistent, group of corporate executives whose chief complaint has been the cost of compliance. No doubt, complying with section 404 costs money. Nor is there doubt, however, that section 404 is working—and that the costs of ignoring weak internal controls would be far greater. Yet smaller companies are where most of the problems historically have existed.

The results of this study indicate that restatements occur twice as often at smaller companies as they do at larger companies. Coinciding with this, material weaknesses are twice as prevalent at smaller companies as they are at larger companies. [Based on material-weakness rate by market capitalization (sources: Glass Lewis; FactSet).] In 2005, the material-weakness rate for companies with market capitalizations under $250 million was 10.2% and the rate by companies with market capitalizations of $2.5 billion or more was 4.7%.] This should not be surprising, given that many small-business issuers do not have designated CFOs. According to a February 2006 report by SME Capital Markets, 251 of the 881 companies that filed Form SB-2 registrations in 2005 did not name a CFO. Others said they relied on part-time CFOs.

The need for effective internal controls at small companies cannot be overstated. In their research, the authors routinely found companies disclosing material weaknesses due to a lack of qualified accounting personnel. In such cases, companies often explain that their accounting departments couldn’t keep up with their operations’ rapid growth. In effect, small companies often grow faster than their internal-control systems can handle.

Restatements and material weaknesses in internal controls. PCAOB Auditing Standard (AS) 2, paragraph 10, defines a “material weakness” in internal controls as a deficiency that results in “more than a remote likelihood” that an error in a company’s financial statements will not be caught. Clearly, without the internal control reviews mandated by section 404, many previously undetected weaknesses would have remained undetected. Section 404 reviews have disclosed thousands of material weaknesses, many of which were directly responsible for financial-reporting errors. As Exhibit 5 shows, more than half of all restatements in 2005 were by companies that disclosed at least one material weakness. (Except where noted, all references to restatements refer only to restatements by U.S. public companies.) Of the 640 restatements last year by companies with disclosed material weaknesses, 42 were by companies that restated twice in 2005. That leaves 598 companies with disclosed material weaknesses that restated in 2005.

Analysis reveals that before late 2004 most restatements were by companies that had not disclosed control weaknesses. This trend shifted during the fourth quarter of 2004, when section 404 work began in earnest. Over the past five quarters (fourth quarter 2004 and all four quarters of 2005), most restatements have been submitted by companies that also disclosed material weaknesses.

Another view is that nearly half of 2005’s restatements were made by companies that have not disclosed material weaknesses. Hence the question: If these companies’ internal controls really had no weaknesses, why did the companies have errors that required restatements? Recall that PCAOB AS 2 defined a material weakness as a deficiency in internal controls that creates “more than a remote likelihood” that a material misstatement in the financial statements will go undetected. If a company restated, then a material misstatement did in fact go undetected, indicating the company’s internal controls must have had at least one weakness.

Often, however, companies get lucky and their weaknesses do not lead to misstatements. By this logic, there should be more companies reporting material weaknesses than restating financials. However, only 640 of the 1,195 restatements in 2005 by U.S. public companies were made by companies that disclosed material weaknesses. Sometimes companies that disclosed the existence of material weaknesses did so belatedly. That is, they first told investors their controls were effective. Then they found errors on their financials and restated, at which point they realized that they must have material weaknesses after all. In these cases, at the same time they filed their restatements, the companies disclosed that their internal controls actually were ineffective.

Some prominent executives and policymakers have complained that the major accounting firms have been going overboard in forcing companies to report material weaknesses. The facts, however, indicate otherwise. If anything, it appears the auditors may have been cutting companies some slack. The authors’ research and analysis found 566 companies that, since SOX section 404 took effect in November 2004, have: 1) restated their financials, and 2) not disclosed any material weaknesses.

Most companies that have restated but did not disclose material weaknesses were nonaccelerated filers. (The authors’ research found 388 of these companies since November 2004.) They were not required to have their internal controls audited independently. Under SOX section 302, which took effect in 2002, these companies’ executives were required to disclose any material weaknesses they found on their own. The authors suspect they would have found many more had they known their outside auditors would be conducting their own independent assessments.

Additional restatement drivers. Today, every accounting firm that audits more than 100 U.S.-listed companies must submit to annual inspections by the PCAOB. These inspections have helped foster a new culture of accountability at these firms. Not long ago, companies often could count on their outside auditors to help them rationalize aggressive—or even simply wrong—accounting practices. Now, audit firms know the PCAOB will be second-guessing any work that appears substandard.

For example, as part of its inaugural inspections of the Big Four, PCAOB inspections forced at least 20 companies to restate their 2003 balance sheets to correct improper debt classifications (“Accounting Overseer Hits the Books,” The Wall Street Journal, September 10, 2004). Since then, PCAOB inspection reports have routinely singled out audit firms for failing to catch material errors that eventually required restatements.

As executive compensation packages continue to soar (up 15% in 2004), and senior officers’ pay remains tied to hitting short-term profit targets, there are strong motives for managing earnings to meet the expectations of analysts and investors. At the same time, with the creation of the PCAOB in 2002, the auditing profession is more highly regulated than ever. These two trends may help explain last year’s historic surge in restatements. Not only do companies continue to give investors bad numbers, but auditors appear to be catching them more frequently, albeit after the fact.

Over time, as companies continue to improve their internal controls, one would expect the number of restatements to eventually decline; perhaps such a trend will be apparent as soon as the 2006 filings are completed. Nevertheless, some executives will try to cook their companies’ books no matter how many rules and laws are passed. As such, financial irregularities will never be eliminated completely.

Restatements, by error category. Investors are more concerned about some restatements than others. Investors may care little if, for example, a company misclassified a short-term certificate of deposit as “other current assets” rather than “cash and cash equivalents.” But they will probably be less forgiving of a revenue-recognition error that overstates sales by 50%. Similarly, a restatement that shifts past sales into future periods may not have the same stock-market impact as a restatement that eliminates a large portion of past sales. The authors’ study classified restatements into 12 categories (see Exhibit 6).

The authors’ review of last year’s 1,195 restatements by U.S. public companies identified 1,796 separately categorized errors. Exhibit 7 shows the number of restatements related to each error category. Expense-recognition errors accounted for nearly 450 restatements, or 25% of all reported errors. This was the most common type of error identified.

More than half of 2005’s expense-recognition errors stemmed from improper lease-accounting practices (see “Increased Clarity in Accounting for Operating Leases,” on page 24 of this issue). In 2005, 249 companies filed restated financials to correct their accounting for leases. Many of these companies also had misclassified items on their cash-flow statements. Even without the wave of lease-accounting fixes, improper expense recognition still would have been a leading cause of restatements in 2005.

The next most commonly restated items were financial-statement misclassifications, followed by equity-related errors. These comprised 18% and 13% of all identified errors, respectively.

Historically, restatements for revenue recognition have resulted in larger drops in market capitalization than any other type of restatements. (See the May 31, 2001, speech by SEC Chief Accountant Lynn E. Turner at USC’s SEC Financial Reporting Institute.) Furthermore, at least before the Enron scandal, revenue overstatements were involved in more than 50% of all accounting frauds. [Fraudulent Financial Reporting—1987–1997: An Analysis of U.S. Public Companies, sponsored by the Committee of Sponsoring Organizations (COSO) of the Treadway Commission, March 1999.] In 2005 the number of restatements correcting revenue-recognition errors increased to 160, from 122 a year earlier. While revenue-recognition errors remained a leading cause of restatements, other categories grew at much faster rates in 2005; nine of the 12 error categories jumped 65% or more from their 2004 levels.

Frequently cited equity-related errors included inappropriate valuations for stock-based compensation and improper accounting for warrants and convertible debt. Restatements in the equity-error category rose 77% in 2005, fueled by an increase in errors related to stock options and convertible instruments. Errors in other comprehensive income (OCI) jumped 198%, the highest percentage increase of any category in 2005. Improper hedge-accounting practices, primarily at financial institutions and service companies, led to 57 of the 122 OCI-related restatements.

Restatements related to acquisitions and investments more than doubled in 2005. Common errors included improper purchase accounting for business combinations. Some companies used the equity method of accounting for investments that they should have consolidated, and vice versa.

From January to September 2006, 145 (17%) of the restatements were for misclassifications with 100 related to cash flow misclassifications and 28 related to hedge accounting. Interestingly, the January 1 through September 30, 2006, review showed an increase in equity-type restatements, with 179 restatements (21%). These appear to have been driven by convertible-debt errors where companies have had to restate the proportions they allocated to debt and equity. Smaller companies continue to restate more often than larger companies. Approximately two-thirds of the 2006 restatements so far were made by companies with a market capitalization of less than $100 million.

Lease-accounting restatements. After 2005’s wave of lease-accounting restatements, one lesson should be drilled into the minds of accountants: Just because everybody’s doing it doesn’t make it right. In a February 2005 open letter to the AICPA, then–SEC Chief Accountant Donald Nicolaisen noted the large number of companies improperly accounting for lease transactions. The letter, which Mr. Nicolaisen issued at the request of the accounting profession, laid out the SEC staff’s view on the correct way to do things—a view that FASB happened to share. As it turned out, hundreds of companies simply hadn’t been following GAAP.

The infractions centered on fairly black-and-white violations of well-established accounting rules. (SFAS 13, Accounting for Leases, was issued in 1976, and FASB Technical Bulletin 85-3, Accounting for Operating Leases with Scheduled Rent Increases, issued in 1985, clarifies the misapplied rules.) Typically, the violations—most of which occurred in the restaurant and retail industries—had the effect of understating rental expenses or improperly keeping lease obligations off the balance sheet. These violations, which had gone on for decades, ultimately led to 249 restatements in 2005 to correct improper lease accounting.

This may sound familiar, because in 1999, FASB Staff Accounting Bulletin (SAB) 99, Materiality, debunked the widely held but incorrect notion that companies could let accounting misstatements slide as long as they fell below certain quantitative benchmarks. (Some accountants used to call this the 5% rule of thumb.) Some companies made one-time adjustments in the fourth quarter of 2004 to correct their lease accounting but avoided restatements by citing immateriality.

Hedge-accounting restatements. Almost a year to the day after the lease-accounting restatement frenzy began, a new but familiar wave of restatements took its place. After an initial surge in March, hedge-accounting restatements started to roll in one after another. Perhaps it was the highly publicized hedge-accounting woes at Fannie Mae that prompted both Fannie Mae’s auditor at the time, KPMG, and other companies to reassess their own practices. Whatever the case, by year end, 57 companies had restated due to hedge-accounting errors. Because the outbreak of hedge-accounting restatements started in late 2005, this issue will probably continue well into the 2006 financial statements.

The 57 hedge-accounting restatements does not include restatements by federal home-loan banks that do not file reports with the SEC, although some of these banks also restated in the wake of irregularities at Fannie Mae and its government-chartered cousin, Freddie Mac.

As with the lease-accounting restatements, the hedge-accounting problems stemmed from companies abiding by a supposed industry norm, notwithstanding that the norm ran counter to GAAP requirements. Once one company has been scolded publicly, everybody else doing the same thing has to face up to the fact that their accounting isn’t GAAP either. Then they, too, must restate.

Stock-option restatements. The adoption of SFAS 123(R), Share-Based Payment, prompted many companies to review their accounting for stock options and other stock-based compensation. As a result, dozens of companies restated their financial statements, even if only their footnote disclosures. Auditors and corporate managers in the past may not have given footnotes their full attention in the past, but they must be audited too. Over the past few years especially, investors began relying heavily on companies’ stock-option footnotes, because they knew the expenses disclosed there would be harbingers of things to come once SFAS 123(R) took effect. The authors’ analysis found 71 stock-options-related restatements in 2005, compared to 39 in 2004.

“Stealth” Restatements

One of Wall Street’s biggest open secrets is that, increasingly, companies are keeping their restatements under the radar by making it difficult for shareholders to find out about them. As a result, many investors perusing companies’ financial reports are surprised to discover restatements that the companies never previously announced; that is, if the investors discover them at all. The authors call these “stealth” restatements.

“Stealth” means that companies restated: 1) without filing an amended quarterly or annual report; 2) without first announcing the restatement in a press release, filed on Form 8-K Item 4.02; and 3) without citing the restatement as the reason for a late quarterly or annual report, disclosed in a Form 12b-25 “NT” filing. The authors use the term “obscure” if a company filed either an amended report or an 8-K, but not both; also for a company that disclosed a restatement in an NT filing but filed neither an 8-K nor an amended report.

Trick 1: Restate, but don’t amend. When a company files a restatement, it typically does so by filing an amended report for the period affected (for example, a Form 10-K/A for a year period, a Form 10-Q/A for a quarter). The “A” alerts investors that something has changed. Upon reading the filing, an investor usually will find an explanatory note describing the reason for the amended filing; in this case, a restatement of previously issued results. While the majority of 2005’s restatements used amended reports, 45% of restatements did not. In 2004, 9% of restatements were filed without amended reports; in 2003, just 5% were.

Many companies avoid using amended filings by restating previous periods in their next regularly scheduled quarterly or annual filings. For example, many companies restated amounts for their 2003 fiscal years in their fiscal 2004 annual reports, filed on Form 10-K. Had these companies instead filed separate, amended annual reports for fiscal 2003, using Form 10-K/A, their restatements would have been far more transparent to investors.

Companies that file restatements without amending their prior filings often leave investors unaware that prior periods have been restated. The authors suspect this is by design. Investors reading a company’s current financial reports simply may not notice the small print. They may dismiss a restatement as relatively minor, because the company was able to tuck it away quietly in the current period’s results.

Trick 2: Restate, but don’t announce. Another way for companies to avoid drawing attention to restatements is fairly obvious: They don’t file press releases announcing them. In 2004, the SEC created a section in the Form 8-K Current Report (commonly used to file press releases) to be used by companies to warn investors that they shouldn’t rely on previously issued financial statement (SEC Release No. 33-8400, Additional Form 8-K Disclosure Requirements and Acceleration of Filing Date, effective August 23, 2004). Since August 23, 2004, companies have been “required” to file Form 8-K, Item 4.02, to announce restatements and alert investors not to rely on the previously issued financial statements affected by the restatement. (See Form 8-K, Current Report General Instructions, Item 4.02.) An Item 4.02 must be filed if: a) “the registrant’s board of directors, a committee of the board of directors or the officers … concludes that any previously issued financial statements, covering one or more years or interim periods for which the registrant is required to provide financial statements … should no longer be relied upon because of an error in such financial statements as addressed in [APB Opinion 20],” or b) “the registrant is advised by, or receives notice from, its independent accountant that disclosure should be made or action should be taken to prevent future reliance on a previously issued audit report or completed interim review related to previously issued financial statements.” The authors put “required” in quotes because there are some big loopholes.

Shortly after the new reporting requirements were issued, the AICPA’s SEC Regulations Committee explicitly asked the SEC if all restatements needed to be reported on Form 8-K pursuant to Item 4.02. The SEC staff said they would support the view of the profession and not require all restatements to be announced in a Form 8-K, Item 4.02. The decision has led to significantly less transparency surrounding the reporting of restatements.

In this case, the rules themselves may be to blame. Where a company’s board, officers, or outside auditor concludes that any previously issued financial statements should not be relied upon, the company has four days to file a report disclosing that such an event has occurred. However, if the company includes that same information in a quarterly or annual report before the four-day period ends, the information need not be repeated in an 8-K.

When that four-day period begins, however, is in the eye of the restater. For example, a company’s auditor and board members may haggle for months about whether a restatement is necessary—before they finally “conclude” that it is. In that case, the conclusion date may fall conveniently within four days of when the company was scheduled to file its next quarterly or annual report. Outsiders can do little to challenge the company’s judgment. And so the restatement announcement gets buried.

Prior to the effective date of Item 4.02, some companies did announce restatements in Form 8-K filings. In 2005, 66% of companies that restated announced their restatements using Form 8-K, Item 4.02. The authors believe this number probably would have been closer to 100% had the SEC not given companies such broad discretion in filing Item 4.02 disclosures.

That said, the revised 8-K reporting requirements have improved transparency for investors. In 2003, just one-third of restatements were announced using 8-K filings; in 2005, two-thirds were. Still, there were 406 restatements in 2005 where, in the authors’ view, the companies did not properly alert investors.

Trick 3: File late, keep quiet, then restate. The majority of companies that restate postpone at least one annual or quarterly report before they complete their restatements. For 64% of 2005’s restatements, the company also was late in filing either a quarterly report or an annual report during the year. This was up from about 54% in both 2003 and 2004.

Assuming they become aware of a company’s intention to restate, investors may interpret a late filing as an indication of the restatement’s severity. It also may signal the depth of the control weaknesses that precipitated the restatement. Companies typically wait to file restatements until they determine the correct amounts and believe they have fixed the problem that caused the restatement in the first place. If a company is in the process of restating—and unable to file financials on time—one would expect it to explain just that in the late-filing notices. This isn’t always the case, however: Many companies fail to provide adequate explanations in their NT filings.

In the authors’ study, of the 191 companies restating in 2005 that didn’t use 8-K forms or amended financial reports to disclose their restatements, 107 filed NT notices before they restated. Of those 107, 76 did not explain in their NT notices that their filings were late because they were busy restating their previous periods.

Stealth restatements—the trifecta. To recap, 535 of 2005’s restatements did not use amended returns. There were 406 restatements that weren’t announced in 8-Ks. As for the 760 restatements that came less than a year after NT notices, the vast majority of those NT notices didn’t cite the companies’ forthcoming restatements as the reason the companies’ filings were late.

About 14% of 2005’s restatements were “stealth” restatements that completed the trifecta. That is, they were: 1) not filed via an amended filing, 2) not announced in an 8-K, and 3) not mentioned in any late-filing notices. The first time that readers of the companies’ filings had the opportunity to learn of these restatements was when the companies filed their regularly scheduled annual or quarterly reports, in which they tucked the restated periods behind their latest periods’ numbers. There were 160 stealth restatements in 2005, up from 13 in 2004 and just two in 2003.

Last year, 21 stealth restatements were due to lease accounting issues. Companies with market capitalizations of less than $75 million accounted for 102 of last year’s 160 stealth restatements. For companies filing restatements, the authors believe the best practice is to file an 8-K, Item 4.02, as soon as they know they will restate, warning investors that they no longer should rely on previously issued financial statements. Companies should also file their restatements using amended filings that explain exactly what was corrected. And if the restatements delay their filings, they should explain so in their late-filing notices.

Restatements and Securities Litigation

Contrary to what one might expect—a corresponding rise in the number of investor lawsuits seeking to recoup losses through class actions—the number of securities class actions has gone down, mainly due to a decline in investor losses tied to such suits. While restatements by U.S. public companies in 2005 set an all-time record of 1,195, securities class actions fell 17% to 176, according to a Cornerstone Research study that also found that 89% of these lawsuits in 2005 alleged misrepresentations in financial documents, compared with 78% a year earlier.

Because not all restatements trigger substantial market-capitalization losses, restatements don’t necessarily lead to securities class actions. Investors are more likely to suffer large losses—and in turn file lawsuits—if a company discloses revenue overstatements, as opposed to, say, a balance-sheet misclassification.

Consider last year’s 249 lease restatements. A few of these restatements may have increased expenses materially. Still, investors perceived most—but not all of them—to be mere technicalities, with little impact on companies’ fundamentals. Such restatements aren’t likely to lead to investor lawsuits when there are no corresponding drops in stock prices.

In its review of 2005 lawsuits, Cornerstone found the three most frequently mentioned accounting allegations were revenue overstatements (40 cases), overstatements of accounts receivable (17 cases), and understatements of liabilities (14 cases). Because these accounting issues are the ones most likely to result in a drop in share price, they are more likely to lead to litigation.

Surprisingly, only five of the 176 shareholder suits filed in 2005 named companies’ outside auditors as defendants, down from eight in 2004, according to Cornerstone. That’s less than one case for each of the six largest accounting firms—Deloitte & Touche, Pricewaterhouse-
Coopers, KPMG, Ernst & Young, Grant Thornton, and BDO Seidman. By comparison, companies audited by these firms filed 782 restatements in 2005.

Auditor Analysis

The first responsibility of auditors is to protect the investing public. Their job is to ensure that companies’ financial statements are presented fairly and comply in all material respects with GAAP. Unfortunately for investors, too often auditors fall short of their objectives.

Exhibit 8 shows that Grant Thornton had the highest restatement rate (number of restatements divided by total number of public companies audited) during 2005, followed by BDO Seidman. Of all the companies audited by Grant Thornton, 12% restated during 2005, compared with 3.8% in 2004. Among the Big Four, KPMG’s 2005 restatement rate was highest: 7.1%; Deloitte & Touche’s 5.6% rate was the lowest.

Assessing what it means for an auditor to have a high or low restatement rate is difficult for outsiders. An accounting firm with a low restatement rate may be providing high-quality audits; or a low rate may signal that the auditor seldom forces companies to correct mistakes. Although a high restatement rate could indicate poor audit quality, it may also indicate that a firm is more willing than others to require corrections.

PricewaterhouseCoopers and Deloitte had the largest volume of restatements in both 2005 and 2004. Both had 200 or more restatements last year. While Deloitte had the lowest restatement rate among the Big Four, its volume of restatements increased by 130% in 2005, the largest increase of the Big Four.

Companies audited by the six largest accounting firms, including Grant Thornton and BDO Seidman, accounted for about two-thirds of 2005’s restatements. The number of restatements at companies audited by other firms more than doubled: 413 in 2005, compared with 189 in 2004.

Audit firms with less than $100 million in annual revenue had the highest restatement rate by far. In 2005, 37.2% of the companies audited by smaller firms restated. The authors view this rate as intolerable. By comparison, the Big Four’s combined rate was 6.1%; the second-tier firms’ combined rate was 8.4%.

Under SOX, annual PCAOB inspections are required only for firms that audit more than 100 companies with U.S.-listed securities. Smaller auditors must be inspected once every three years. The authors’ study, which found the restatement rate for smaller firms extraordinarily high, demonstrates that triennial inspections may not be frequent enough for some firms.

While companies audited by smaller firms restate more frequently, companies audited by the Big Four still account for the largest volume of restatements. In 2005, 706 restatements were at companies audited by the Big Four, compared with 389 at companies audited by small firms. The number of Big Four restatements increased 81% during 2005, while the number of restatements at small firms jumped 114%.

The auditor-turnover rate for companies with restatements was twice as high as the overall auditor-turnover rate. From 2003 to 2005, the average turnover rate for all companies was about 12%. The average auditor-turnover rate for companies with restatements during the same three-year period was about 24%.

Smaller Companies Restate More Often

More than 11% of U.S. public companies with market capitalization of less than $250 million restated during 2005, compared with about 6% of companies with market capitalizations of $2.5 billion or more. Restatement rates increased across all size categories in 2005; the rates doubled among companies with market capitalizations of more than $250 million. About one of every 19 companies in the largest category ($7.5 billion or more) restated financials in 2005, compared with just one of every 46 in 2004.

Even with the high growth in restatements by larger companies, what stands out is that one of every eight small companies (market capitalization under $75 million) restated in 2005. More than half of last year’s restatements were filed by the smallest companies, which registered 604 restatements. The number of restatements declines from there as market capitalization rises.

The trend in 2004 was clear: The smaller the market capitalization or annual revenue, the more likely a company was to restate. However, the trend in 2005 was more mixed. Higher annual revenues in 2005 didn’t necessarily translate into lower restatement rates, although higher market capitalizations did.

Sarbanes-Oxley Was Not an Overreaction

Financial statement restatements continue to rise. But according to a relative handful of senior executives and policy makers, the big problem isn’t that investors got bad information the first time; rather, they contend that Congress overreacted when it passed SOX and that, for the good of global capitalism, Congress should now relax the scrutiny it imposed on corporate boards and executives after the collapses of Enron and WorldCom.

After surveying the accounting mishaps and do-overs that were reported in 2005 and first nine months of 2006, the authors couldn’t disagree more. It’s precisely because of the heightened auditing standards mandated by Sarbanes-Oxley that investors are finally getting a true sense of how much work remains to be done before they can feel confident about the accuracy of the financial statements prepared by corporate managers.

Lynn E. Turner, CPA, is managing director of research at Glass Lewis & Co., LLC, and senior advisor to Kroll, Inc.
Thomas R. Weirich, PhD, CPA
, is a professor of accounting at Central Michigan University, Mt. Pleasant, Mich.