Delays and mistakes in SEC financial reporting can have far-reaching consequences for companies and even for their shareholders. The impacts may include:
- SEC Enforcement reviews, actions, and penalties
- An SEC comment letter, demanding a time-consuming and expensive response
- A required filing amendment, including financial restatements
- A drop in the company’s stock price
- Complications with the company’s stock-exchange listing
- Unfavorable scrutiny by analysts
- Loss of short form (S-3) eligibility and well-known seasoned issuer (WKSI) status
A study reported by The Wall Street Journal in September 2018 shows an uptick in corporate financial restatements during the first half of 2018, the first increase in a decade. [See Companies Are Finding More Accounting Flubs, by Nishant Mohan, Sept. 20, 2018.] The article quotes an accounting expert who observes that the errors prompting restatements “can be anything from a misapplication of accounting principles to an error in inputs in accounting software or an error in [Microsoft] Excel schedules.” Those who should have caught the mistake usually feel the consequences, “particularly...the CFO and the controller.” Notorious examples of data errors in SEC filings have occurred even among high-profile companies (Netflix, for example, was adversely affected by a single data error).
Late filings often cause the stock price to drop
A working paper published by New York University School of Law offers jolting observations about the major real-world impacts of financial-reporting mistakes. SEC Filings, Regulatory Deadlines, and Capital Market Consequences, by business professors Eli Bartov (New York University) and Yaniv Konchitchki (University of California, Berkeley) is also summarized in Columbia Law School’s CLS Blue Sky Blog.
The researchers examined the effects that late quarterly and annual filings had on more than 2,000 companies which had never previously delayed a filing. They found that accounting issues are the most common reason for late filings and often determine the severity of the impact. Among the sampled companies, delayed filings of Forms 10-Q and 10-K were an average of 41 days late when accounting was involved—more than three times later than the average delay because of corporate events (13 days) or “uncertain” reasons (11 days).
Perhaps most alarmingly, the researchers discovered that a company’s stock price often tumbles as soon as the filer indicates the financial statement will be late. When Form 10-K or 10-Q is delayed, SEC Rule 12b-25 requires the company to file Form NT (for “non-timely”). This provides a one-time grace period of five days for Form 10-Q and 15 days for Form 10-K. Among the studied companies, announcements of tardy 10-Q filings caused an average stock-price drop of almost 3% and about 2% for late 10-Ks. Evidently, it does not matter if corporate management promises that the statement will be filed by the new deadline—the stock price still falls, sometimes significantly. The authors also note that companies are twice as likely to exceed the grace period deadline for 10-Qs (51% of the time) as they are for 10-Ks (25%).
Investors interpret amended filings as signs of trouble
Audit Analytics reported a variety of reasons for filing amendments (on Form 10-K/A), including accounting errors and mistakes such as missing signatures and exhibits. [See Amended Filings: What’s Missing in 10-Ks?.] “In the eyes of investors and stakeholders, amended filings can sometimes be perceived as a red flag,” the firm warns. “After all, in an effective control environment, registrants should presumably file full and reliable information on the first attempt. So when a company has to amend a filing—especially a periodic report like the 10-Q or 10-K—it should raise at least some concerns.”
Among the many reasons that Audit Analytics found for companies’ required filing amendments on Form 10-K/A:
- Omissions in proxies (e.g., Part III and Item 10-14)
- Missing exhibits or signatures
- Problems in the financial statement of a subsidiary
- A full financial restatement
- XBRL-related matters
The firm notes that these problems “could be indicators of process weaknesses or other underlying issues” and pointedly advises investors to take note of “the reason that an amended filing was required, and whether key information was omitted from the original annual reports.”
SEC comment letters influence investor behavior
Simply receiving an SEC comment letter on a financial filing can have a detrimental effect on the company’s market performance, according to researchers. [See The Power of Words in Capital Markets: SEC Comment Letters on Foreign Issuers and the Impact of Domestic Enforcement, by Daniel Giamouridis, Bank of America Merrill Lynch; Kleopatra Koulikidou International Hellenic University; and Stergios Leventis, International Hellenic University.]
Negative tone from SEC staff will trouble investors
The researchers developed a “bag-of-words dictionary” to assess the tone in SEC comment letters issued to foreign companies listed on US stock exchanges. They discovered that SEC staff criticism (which they term “negative-tone regulatory language”) of mistakes in financial reporting significantly affects investor behavior, for both US GAAP and non-US GAAP filers. Even if a company’s SEC filings are on time and do not require restatements, errors prompting an SEC comment letter can still hurt the stock price.
XBRL exposes issues to analysts and SEC staff
XBRL-tagged data enables market analysts and the SEC to perform their jobs more effectively. It gives the ability to process and understand the “as filed” data more quickly and comprehensively than was possible with the EDGAR data submitted before the XBRL mandate. Use of XBRL data by investors and analysts is now mainstream, as data providers increasingly incorporate XBRL-tagged disclosures into their feeds. Investors rely on financial information delivered swiftly by XBRL, whether in XBRL file sets from the SEC website or through information systems of vendors, such as Calcbench and idaciti, that use XBRL data.
Moreover, companies no longer have an informational advantage over the SEC, Willkie Farr & Gallagher attorney Elizabeth Gray noted at a Toppan Merrill webinar. With structured data and other technology-driven sources of information, the SEC knows much more about companies than ever before and can independently corroborate or disprove filers’ statements. If there are issues in a company’s disclosures, XBRL data provides the SEC staff the means to identify those areas.
Consequently, it is crucial for an SEC registrant to submit high-quality XBRL tagging so that the company’s reported financial statements are properly interpreted by market analysts and understood by investors. Improper tagging of the XBRL-formatted financial statements, of which the company is often unaware, will convey the wrong information and result in unintended consequences.
No cavalier approach: XBRL errors can trigger a review
Of course, error-free XBRL is also vital for SEC compliance—and in more ways than you might realize. XBRL is a key tool not just in the Division of Corporation Finance but also in the Enforcement Division. Errors in XBRL can trigger closer scrutiny by SEC staff of a company’s financials, resulting in further investigation and inquiries directed to the company.
In short, with XBRL-tagged data, the SEC and the market are able to identify areas of concern promptly and more accurately. The modern, sophisticated information environment no longer tolerates a cavalier approach to XBRL. “SEC staff and market analysts are increasingly utilizing tools dependent on XBRL and other structured data sets to analyze data automatically, resulting in the need for SEC registrants to submit XBRL that is properly tagged—including the detailed footnote disclosures,” says Jennifer Froberg, SEC analyst at Toppan Merrill. Filers must know how to execute XBRL tagging accurately or seek outside providers such as Toppan Merrill to ensure error-free filings.
SEC technology uses XBRL for deeper review
The importance of accurate XBRL is reinforced by looking at the advanced technology that the SEC staff is now using to mine and analyze the XBRL-tagged data in SEC filings. Howard Kaplan is a data analyst in the Enforcement Division’s Center for Risk and Quantitative Analytics. During a Toppan Merrill webinar, he explained the SEC’s use of structured data, such as XBRL. “Structured data is baked into the majority of our work,” he asserted. “The majority of the cases that the SEC brings, as well as the many cases that never get brought, are predicated on some analytics using structured and, to some degree, unstructured data.”
The data analysts give special scrutiny to companies that submit XBRL filings which contain errors. Once XBRL errors become evident, the team starts to dig deeper. “We often come across things such as negative fact values, data-entry errors, use of extensions for common line items,” Mr. Kaplan stated during Toppan Merrill’s webinar. “The analyses we do will be more focused when we find anomalies or poor data quality or poor submissions on the XBRL side.”
The SEC routinely uses structured data in enforcement cases, including those that involve disclosure and accounting violations. The XBRL data is combined with other data sets, such as market data, unstructured data, and information from websites and social media, to paint a detailed, panoramic picture. “It also positions us well for things like the consolidated audit trail,” Mr. Kaplan added.
Enforcement cases that involve accounting and disclosure take longer to bring than others, but with structured data, the SEC can be much more proactive. For example, the SEC staff can swiftly identify an event that should trigger a Form 8-K filing, one that was reported in a periodic filing but not disclosed in an 8-K. Issues that frequently arise in these investigations include revenue recognition, different kinds of accrual issues, and changes in officers and directors. “In every type of 8-K event,” said Mr. Kaplan, “there is usually a connection back to the XBRL filing or 10-K/10-Q in general.”
Filings with XBRL errors go into a special bucket for review
After finding matters that should have been reported on Form 8-K but were not, the data analyst submits the discovery to an enforcement team. “In our screening process,” he warned, “the filings with unnecessary extensions or XBRL errors go into a bucket, where they’re looked at more closely. Often, nothing is there. But it is in a company’s interest not to go into that bucket to begin with.”
Textual analysis of filings
It is not just numbers that the SEC staff is analyzing. Increasingly, the SEC is using textual analytics to probe a filer’s words. One area of inquiry there is “sentiment analysis,” a process of computationally identifying and categorizing opinions expressed in a text disclosure to determine the filer’s attitude toward the topic in question. “Textual-analytical engines are wildly more effective at analyzing structured disclosures when they know what they are looking at, versus attempting to consume and analyze an entire unstructured report,” Mike Willis, Assistant Director in the SEC’s Office of Structured Disclosure suggested in a recent issue of Dimensions.
Scott Bauguess, the Deputy Director and Deputy Chief Economist in the SEC’s Division of Economic and Risk Analysis, discussed the SEC’s analytical capabilities in a speech he gave earlier this year, The Role of Machine Readability in an AI World. In a subsequent interview in the June/July 2018 issue of Dimensions, he explained that XBRL block tagging and detail tagging provide great value for text analysis in the footnotes. “A lot of the important information in a 10-K is tucked into the footnotes and the specialized disclosures,” he observed. “Our ability to extract those footnotes systematically and organize them lets us apply the new technology, whether it is machine-learning text analytics on the footnotes or extracting certain information, such as rates of return, and we can get that information very quickly.” Significantly, he added, with XBRL the SEC staff can now instantly pick up changes in footnotes from quarter to quarter:
Often the examination and disclosure staff wants to know what has changed. We have tools that will redline changes in footnotes from period to period so that an examiner can see what has changed. That is far easier and more powerful than reading one footnote and then going to the other and trying to figure out exactly what is different. The redline helps you understand much more quickly what has changed with that entity.
Fixing XBRL errors: human review is essential
XBRL errors can be conveniently grouped into two categories: mistakes that can be found by software, and errors that are beyond the reach of computers and warrant human review. For those errors that can be detected by software, the baseline use of software to check for mistakes is essential. Good software can check specific items and show where, how, and why the XBRL tagging is not in compliance with the SEC rules.
The SEC put the spotlight on automated checking of XBRL and the accuracy of tagging in its EDGAR Release 18.3 (effective from Oct. 1, 2018), which includes new types of XBRL validation checks. The new error types, which are flagged as warnings (not suspensions), occur when XBRL submissions contain one or more errors of these types:
- US GAAP and IFRS numeric reporting items tagged incorrectly as negative
- Deprecated tags
- Redundant custom axis tags that are already provided for in the US GAAP taxonomy
As Lou Rohman, Vice President of XBRL Services for Toppan Merrill, explains in his blog post: “[P]roviding error messages directly to the filer at the time of the test submission will raise awareness and will likely result in filers correcting the errors prior to live filing, consequently improving their data quality … I expect the number and significance of the items [to be checked] will grow in the not-too-distant future.”
XBRL US warns that in this updated version of the EDGAR Filer Manual, “the Commission is taking note of the quality of the XBRL data submitted by SEC filers and is bringing the importance of data quality to the attention of filers.”
XBRL errors that only a person can detect are as frequent, severe, and painful as the errors that software can find
However, for those errors that cannot be detected by automation, human expertise is required to prepare and review the XBRL files properly. XBRL mistakes that are detectable only by a person can be as frequent, severe, and painful as the errors that software can find. For example, new rules for lease accounting will take effect in 2019. IFRS and US GAAP categorize and tag lease information differently. Filers need expert guidance to implement this and other new financial-reporting disclosures and their related XBRL structures.
Toppan Merrill’s unparalleled XBRL team routinely spots XBRL errors that cannot be detected by software in the financial statements of SEC registrants. Unfortunately, these errors often significantly misreport a registrant’s financial disclosures. Whether from inside or outside the company, the XBRL preparer and reviewer should have a deep understanding of accounting disclosures, experience with the XBRL specification, expertise with the EDGAR Filer Manual, familiarity with the US GAAP or IFRS taxonomy, and knowledge of the XBRL US Data Quality Committee rules.
“Errors that cannot be detected by software really require the involvement of knowledgeable XBRL individuals,” Mr. Rohman emphasizes. “That’s a key point of getting the XBRL right. By knowledgeable, I mean individuals who understand the SEC rules, the accounting disclosures, the intricacies of the XBRL itself, and then are able to apply that experience to the XBRL document.”