The SEC’s Newest Tool is Headed to the Front

In the ongoing battle against financial reporting fraud, the Securities and Exchange Commission (SEC) is employing technology in a way that is making waves in Wall Street circles – enough so that, according to a recent Forbes article, a key tool in the SEC’s approach has been given a nickname: RoboCop.

August 2013 

By ACFE Staff 

 

In the ongoing battle against financial reporting fraud, the Securities and Exchange Commission (SEC) is employing technology in a way that is making waves in Wall Street circles — enough so that, according to a recent Forbes article, a key tool in the SEC’s approach has been given a nickname: RoboCop. The obvious image (at least for those familiar with the futuristic movie franchise of the same name) is a steel plated, heavily-armed crime fighter who patrols the mean streets of Detroit in a bleak struggle for justice.

 

Computers analyzing financial statements for accounting anomalies does not, admittedly, inspire such drama. But in the effort to detect more fraud, the new weapon that the SEC is calling its Accounting Quality Model (AQM) has real potential to be a game-changer. For fraud examiners, data analysis and other technology-driven means used by AQM will be familiar and, for some, old hat. But employed by the SEC as a resource for reviewing potentially thousands of corporate filers, it stands to reason that leads, audits and enforcement actions could dramatically increase in the near future.

 

Why Now? 

In 2012, only 11 percent of the SEC’s enforcement actions resulted from accounting and financial-disclosure fraud, according to the Forbes piece. SEC Chairman Mary Jo White has signaled a desire to step up such actions. According to an SEC press release posted in July, the AQM is an element of their Division of Enforcement's “ongoing efforts to concentrate resources on high-risk areas of the market and bring cutting-edge technology and analytical capacity to bear in its investigations.”  

 

Notably, the other main elements are the SEC’s Financial Reporting and Audit Task Force, dedicated to detecting fraudulent or improper financial reporting; and the Microcap Fraud Task Force, targeting abusive trading and fraudulent conduct in securities issued by microcap companies.  

 

The press release stresses that the technology and analytics tools are meant to make the SEC’s hands in the field more effective — in other words, RoboCop is not replacing the street cop in terms of real investigative work. As is the case for fraud examiners at large, real people are needed to interpret the findings from quantitative analytics and provide follow-up investigation to determine true facts. The implications are clear, however — the potential for broad, first-line detection of red flags could fill a gap where incidences of fraud aren’t tipped by whistleblowers. 

 

AQM: Extending the Traditional Approach 

Last December, the SEC’s Craig M. Lewis, Chief Economist and Director, Division of Risk, Strategy, and Financial Innovation (RSFI), described in some detail the way that AQM will be implemented. His remarks included a bit of background:

 

First, the premise of all models that seek to identify earnings management is that firms have strong incentives to manage earnings. The evidence broadly follows two strains: One, investors respond to earnings announcements and, two, earnings management by the firm influences market information about the firm’s future performance and investment prospects.  

  

Second, we need to understand generally where it is possible to discern the effects of earnings management. Typically, they manifest in the discretionary choices that management can make under GAAP when reporting its financials. In accounting jargon, total accruals are the difference between free cash flows and income before extraordinary items. It is the difference between what accountants recognize as revenue and expenses and the actual cash flows available to shareholders. We can decompose total accruals into two broad categories: discretionary accruals and non-discretionary accruals.  Non-discretionary accruals are accounting adjustments made in strict adherence to GAAP and are relatively objective. Discretionary accruals however, may be subjective and require the preparer to exercise considerable accounting judgment. As is generally recognized, this influence over the potential accrual values can allow for opportunities to, for example, smooth income and therefore, manage earnings most aggressively. 

 

Lewis goes on to explain how AQM works as a complement to traditional models – he cites the “Jones” (created in 1991) or “modified Jones” models. One of the chief problems with these models has been the resultant false-positives. To counter this issue, the AQM extends the traditional approach “by allowing discretionary accrual factors to be part of the estimation,” according to Lewis:

  

Specifically, we take filings information across all registrants and estimate total accruals as a function of a large set of factors that are proxies for discretionary and non-discretionary components. Further, we decompose the discretionary component into factors that fall into one of two groups: factors that indicate earnings management or factors that induce earnings management. Discretionary accruals are calculated from the model estimates and then used to screen firms that appear to be managing earnings most aggressively. 

 

  

What it Means (in Plain English) 

There is more explanation from Lewis along these lines, and if you’ve followed the accounting jargon thus far, you’ll find the conclusion of his remarks to be of interest. If not, then here is the important part: according to BakerHostetler associate Francesca Harker (writing in Forbes), “Based on a comparison with the filings of companies in the filer’s industry peer group, the AQM produces a score for each filing, assessing the likelihood that fraudulent activities are occurring.” Harker further observes:

  

The results of RoboCop’s analysis will likely become the basis for enforcement scheduling and direction of resources in the near future. A filing’s risk score will determine whether a filing is given a quick, unsuspecting review, or whether it is thoroughly dissected by an SEC exam team, possibly leading to an expensive audit. The SEC has also said it plans to use the risk scores as a means of corroborating (or invalidating) the approximately 30,000 tips, complaints, and referrals submissions it estimates will be received each year through its Electronic Data Collection Systems or completed forms TCR. 

 

Among the advice Harker has for corporations to avoid falling into the AQM’s crosshairs: Check your work, use accounting policies consistent with your peers, stick with one auditor, reduce off-balance sheet transactions, use conservative decision-making regarding discretionary accruals and be prepared to respond to SEC inquiries.

 

Word Games 

One of the most intriguing potential elements of AQM is an approach to analyze language used in specific areas of corporations’ annual reports. As reported in May by The Wall Street Journal, word choices used in the "management's discussion and analysis" section can “reveal warning signs of earnings manipulation.” Lewis told the Journal that companies that bend or break the rules are often playing a “word shell game.” The Journal further reports:

  

Such companies try to "deflect attention from a core problem by talking a lot more about a benign" issue than their competitors, while "underreporting important risks." 

 

Officials told the Journal that if the word-analysis program works, it will be added to the AQM. No indication as yet on how or when its effectiveness will be determined, but language analysis is something organizations are already doing on their own to detect and prevent fraud. Blogging for Forbes at the 24^th Annual ACFE Global Fraud Conference, Walt Pavlo interviewed presenter Vince Walden, CFE, on how companies are using tools not only to detect fraud – but to predict where it may occur. Included in a battery of new measures is document analysis, including a review of communications (such as emails) for key words. Walden describes how this can be used in combination with other data:

  

“It is an evolving science of analyzing data and looking for patterns and anomalies… we have models that can look at travel and expenses of an employee, tagged to sales activity with a certain customer, tagged to late working hours, tagged to third-party email communications, tagged to various email aliases and then we get a result that tells us a story.” 

 

The actual implementation of word analysis in the AQM would be somewhat different – after all, the wording and phrasing it would evaluate will have gone through proofing and executive approval, as opposed to what is found in a hastily written email – but the principals are the same. As Harkin wrote in Forbes, “through a study of past fraudulent filings, analysts at RSFI have developed lists of words and phrasing choices which have been common amongst fraudulent filers in the past.” 

 

Looking Ahead 

Taken as a whole, the AQM would appear to signal a move forward in Mary Jo White’s pledge to step up SEC accounting enforcement. The true measure will come next year, and the year after, when we discuss the number of enforcement actions resulting from the work of this new RoboCop. The hope is that the tool operates the way its creators say that it should – and SEC investigators can carry the ball from there.