A model for fighting fraud; SEC developing software “to sift language in financial reports for clues that executives might be misstating results”

Saturday June 8, 2013

A model for fighting fraud

Optimistically Cautious by ERROL OH

WHAT if the regulators can extract information from a listed company’s annual report and feed it into a computer program to find out if there’s probably some financial sleight of hand going on? That sounds like a tremendous step forward for capital market supervision and enforcement. It also seems a bit far-fetched. But maybe it’s not. The Wall Street Journal reported on May 29 that the US Securities and Exchange Commission (SEC) was renewing its focus on accounting fraud and other problems relating to financial disclosures. The newspaper quoted agency officials as saying the SEC was already developing software “to sift language in financial reports for clues that executives might be misstating results”. The program, according to the article, would analyse the annual report section (usually called the management’s discussion and analysis) in which the companies talk in detail about their performance and prospects. Certain word choices, readers are told, may be red flags that warn of earnings manipulation. One SEC official said tests to determine if the analysis would have sniffed out previous accounting frauds “look very promising”. Here’s a couple of paragraphs from the story: “Firms that bend or break accounting rules tend to play a word shell game,’ said Craig Lewis, the SEC’s chief economist and head of the division developing the model. “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’.”

Getting comments from Lewis in this context is indeed appropriate. His division is called the Division of Risk, Strategy, and Financial Innovation, and the model it’s working on is called the Accounting Quality Model (AQM).

Officials quoted in the WSJ article said if the word-analysis program worked, it would be added to the AQM, which SEC enforcement staff started using recently.

The newspaper added: “The model trawls data from nearly 9,000 publicly traded companies. A similar computer-powered search for unusual performance patterns at hedge funds has led to seven enforcement actions in recent years.”

The AQM appears to be a fascinating beast, but the news report reveals nothing else about it. To know more, the starting point should be a speech Lewis gave at a gathering of senior financial executives in December.

He describes the AQM as “a model that allows us to discern whether a registrant’s financial statements stick out from the pack, while taking into account the contemporaneous attributes of that pack”.

He says: “The goal is to facilitate comparison across firms within their industry while accounting for and illustrating industry differences as well.”

But we’re still no closer to knowing how the model works. For that, Lewis has to first introduce his audience to the phrase “earnings management”. That’s a euphemism for the aggressive use of discretionary accounting choices in manipulative or even potentially fraudulent accounting practices.

He points out that the phrase is broad enough to capture both aggressive accounting practices that fall within GAAP (generally accepted accounting principles) and fraudulent accounting practices that violate GAAP.

Also, to understand the model better, we need to accept that companies have strong incentives to manage earnings, and we have to grasp the accounting concept of accruals.

Lewis explains that total accruals are the difference between what accountants recognise as revenue and expenses, and the actual cash flows available to shareholders. There are two categories: discretionary accruals and non-discretionary accruals.

The former are accounting adjustments made in strict adherence to GAAP and are relatively objective, he says.

On the other hand, discretionary accruals may be subjective and require the preparer to “exercise considerable accounting judgment”. “As is generally recognised, this influence over the potential accrual values can allow for opportunities to, for example, smooth income and therefore, manage earnings,” adds Lewis.

Discretionary accruals that stick out like a sore thumb can be “a powerful indicator of attempts to manage earnings”. He says: “The trick is to identify those outliers.”

There are already other models for detecting earnings management, but according to Lewis, the SEC’s AQM goes further by being more specific about the discretionary accruals. This is possible partly because the model incorporates the SEC staff experiences and knowledge.

Lewis ends the talk by tempering expectations. “The AQM I’ve been talking about today represents an additional approach to estimating discretionary accruals and identifying outliers, which can, in turn, assist in the effort to identify earnings management,” he says.

“This is not to say that we’ve built a model that can detect fraud’. Far from it. Rather, we hope to provide one more tool that the already sophisticated staff of the SEC can use in its efforts to ensure high-quality financial statements.”

This bit of downplaying is meant to wipe out any notion that the SEC has found a magic bullet. But any addition to the fraud-busters’ arsenal especially one that looks formidable and well-conceived is a welcome thing. At least, it will make the crooks work harder to beat the system.

> Executive editor Errol Oh wonders if the tired corporate phrase “cautiously optimistic” is a trigger in the SEC’s word-analysis software.