The SEC’s renewed focus on accounting fraud and how it is leveraging technology to tip the scales

July 2015  |  SPECIAL REPORT: WHITE-COLLAR CRIME

Financier Worldwide Magazine

July 2015 Issue

The Securities and Exchange Commission (SEC) has a rather daunting three part mission, to “protect investors; maintain fair, orderly, and efficient markets; and facilitate capital formation”. The mission is especially difficult when you consider the SEC has approximately 4700 full time employees who have responsibility for close to 35,000 entities including 9000 reporting companies. According to a recent Harvard Business School working paper, these reporting companies make an average of 3000 filings per day.

In light of these numbers, the SEC has turned to technology to help execute its mission. This article provides an update on the SEC’s renewed focus on financial statement fraud and highlights several areas where the SEC has looked to technology, data analytics and computer science for assistance. It has a particular focus on the much publicised Accounting Quality Model (AQM) used to identify accounting fraud as well as the Corporate Issuer Risk Assessment (CIRA) program that has expanded the SEC’s efforts in this area.

SEC enforcement developments

The sheer number of corporate filings necessitates that the SEC look for new ways to efficiently identify fraudulent financial statements. In financial year 2012, the SEC had 79 financial fraud/issuer disclosure actions, a historically low number. By comparison, 2014 shows an uptick with 99 financial fraud/issuer actions with monetary remedies across all categories topping $4.16bn. Why the change?

An SEC press release from 2 July 2013 outlined multiple new initiatives to combat financial reporting fraud, including the creation of a Financial Reporting and Audit Task Force (Task Force) within the Division of Enforcement to assist with enforcements related to suspected accounting and disclosure fraud. The Task Force includes accounting and legal personnel with access to accountants, economists and attorneys throughout the agency.

At SEC Speaks 2014, Chair Mary Jo White reiterated that the Task Force would look for patterns of conduct indicative of financial fraud in areas such as revenue recognition, asset valuations and management estimates. At that time, the Task Force had already generated several investigations. The July press release also announced the creation of the Center for Risk and Quantitative Analytics (CRQA) to serve as an “analytical hub and source of information about characteristics and patterns indicative of possible fraud”. Both the Task Force and the CRQA are drawing on new analytical and technological tools being developed at the SEC, such as the AQM/CIRA.

The Accounting Quality Model (AQM)

Financial data is one thing that the SEC does not lack. AQM is an analytical attempt by the SEC to review that data. In a number of speeches, Craig Lewis, the former Chief Economist for the Division of Economic and Risk Analysis (DERA), describes AQM as a customised analytical tool designed to detect whether a registrant’s financial disclosures differ greatly from its industry peers. The SEC believes that if a registrant has financial metrics that are outliers for its peer group, that finding could indicate a need for additional scrutiny. The SEC requirement that financial data filed with the SEC be tagged with an Extensible Business Reporting Language (XBRL) format normalises the data and enables AQM to make comparisons among filers.

The SEC initially designed AQM primarily for the Division of Corporation Finance (Corp Fin) to improve financial disclosures in its registrant filing reviews. As noted by Mr Lewis at the XBRL US National Conference, the model would both risk rank filings and direct Corp Fin reviewers to areas of initial focus. In addition, the Division of Enforcement uses AQM to help identify financial fraud.

How AQM works

To understand how AQM works, it is first necessary to understand discretionary and non-discretionary accruals, and how fraud can sometimes find its way into discretionary accruals. Companies record accruals to recognise unpaid liabilities. Accruals represent one component of the difference between accounting income and cash flows, and can either be non-discretionary or discretionary. Non-discretionary accruals are standard accounting entries at the end of an accounting period, such as ‘accrued salary’ for wages earned but not yet paid, or ‘accrued rent’ for unpaid lease obligations. Discretionary accruals also recognise liabilities, but involve management discretion. Examples include accruals for uncollectible accounts receivables (bad debts) or contingencies, such as potential exposure on litigation. These accruals can also require management judgment. Pressure to meet earnings targets can cloud that judgment and lead to improper accounting.

According to Mr Lewis, similar to other earnings management detection models, AQM focuses on discretionary accruals and their subjective nature. Historically, these analyses modelled a company’s non-discretionary accruals, and then used the remaining accruals (the unexplained portion) as a measure of management discretion. Mr Lewis describes AQM as an attempt to go beyond traditional earnings management detection analyses by modelling both non-discretionary accruals and discretionary accruals. It further divides discretionary accruals into two groups: (i) factors that indicate earnings management (i.e., factors directly associated with earnings management); and (ii) factors that induce earnings management (i.e., factors that are inclined to incentivise earnings management). He also noted that this approach is designed to minimise false positives and create a working list of outlier registrants.

In a number of speeches, Mr Lewis and former chairman Elisse Walter went into some detail on indicators and inducers. Cited examples of accounting policies that might indicate earnings management include accounting policies that encourage off-balance sheet transactions, or policies that encourage high book earnings while conversely selecting tax strategies that minimise taxable income. Risk inducers, on the other hand, include scenarios where a registrant is losing market share or has lower profitability than its peers. AQM creates a score that ranks filers on the basis of risk and identifies outliers. The output also highlights the factors that make the filer an outlier, thereby providing direction for SEC reviewers.

Risk assessment textual analysis in AQM

According to Mr Lewis, DERA is looking to further enhance AQM by including a textual analysis of registrant filings. The SEC analysed past enforcement actions and found that fraudulent filers often use certain terminology or word emphasis – for example, they might highlight benign or unimportant aspects of a filing, or gloss over key industry risks. Based on these and the aforementioned factors, the model would then generate a numerical risk-ranked score. Again, the purpose of this textual model is not solely for enforcement, but to help Corp Fin review registrant filings.

Corporate Issuer Risk Assessment (CIRA)

Mark Flannery, current chief economist for DERA, has hinted at the expansion of AQM into a broader risk assessment. In February 2015, Mr Flannery indicated that the original focus of AQM is now only one of more than over 100 custom metrics provided by DERA. For example, CIRA has the ability to see inventory build-up relative to sales. Enforcement may use this metric, along with other risk factors, as a possible indicator of fraud (i.e., the company might be aggressive in its accounting).

Takeaway

The SEC continues to adopt advanced technology, mathematics and computer science to both improve corporate financial disclosures and identify and investigate potential accounting fraud. The SEC’s new tools require corporate financial executives to remain diligent in evaluating accounting accruals, establishing effective internal controls and ensuring proper XBRL labelling for financial disclosures. In addition, the SEC has disclosed a number of risk inducers and risk indicators that will most likely be flagged by AQM/CIRA. As always, filers should understand how their financials differ from their peers and be readily prepared to explain those differences to the SEC.

 

Rick Ostiller is a managing director and John Stark is a director at Navigant Consulting Inc. Mr Ostiller can be contacted on +1 (650) 849 1171 or by email: rostiller@navigant.com. Mr Stark can be contacted on +1 (415) 399 2142 or by email: jstark@navigant.com.

© Financier Worldwide

 

BY

Rick Ostiller and John Stark

Navigant Consulting Inc.

©2001-2024 Financier Worldwide Ltd. All rights reserved. Any statements expressed on this website are understood to be general opinions and should not be relied upon as legal, financial or any other form of professional advice. Opinions expressed do not necessarily represent the views of the authors’ current or previous employers, or clients. The publisher, authors and authors' firms are not responsible for any loss third parties may suffer in connection with information or materials presented on this website, or use of any such information or materials by any third parties.