Although often overlooked, accruals recorded in financial statements offer an opportunity to manipulate financial information and earnings management in order to promote a healthier image of the company to its investors. Management of earnings from period to period known as “income-smoothing” does not necessarily rely on misstatements that constitute fraud, but more on various interpretations of GAAP. Yet, this practice may meet the definition of fraud when the manipulation is intentional and executed to deliberately misrepresent earnings to investors. In other words, there is a fine line between income smoothing and fraud. Most investors are unaware of this practice and this technique does not necessarily provide particularly accurate financial information from period to period. What should be very important to investors though is the relationship between income smoothing, net income and future cash flows. In the case of XYZ, Inc., manipulating accruals allowed the company to misrepresent its net worth to its shareholders.
In order to understand how this works, one must first understand the true definition of an accrual – anything other than cash. Traditional thinking considers accruals in the liability section of the financial statement such as accrued payroll, accrued income taxes, other accrued liabilities, accounts payable and notes payable. Yet, accruals exist in the asset section as well, including such accounts as accounts receivable, warranty reserves, allowance for bad debts, goodwill, intangibles and inventory reserves. Some of these accruals, such as accrued bonuses, allowance for bad debts, allowances for loan losses, warranty or inventory reserves are based on management’s estimates and provide excellent sources for financial statement manipulation.
Measuring accruals determine their effects on current income and future cash flows and can identify possible manipulation of earnings or possible “income-smoothing” of the financial statements. There are three different models for measuring the impact of accruals in financial statements: Dechow-Dichev Accrual Quality, Sloan’s Accruals and Jones Nondiscretionary Accruals. Each model is unique and allows a forensic approach to measuring accruals in the financial statements to determine whether any unusual variations or anomalies exist. Generally, one should use all three models to determine whether unusual variations exist and reduce the possibility of “false positives” in the testing. Sloan’s Accruals is especially adept for focusing on specific periods while all three models may be used for monthly, quarterly or annual analyses of financial information. Abbreviations used for the following formulas include:
Current year = cy
Prior year = py
Working Capital = WC
Change = ∆
Total Assets = TA
Revenue = Rev
Gross Property, Plant and Equipment = PPE
The Dechow-Dichev Accrual Model discussed in their paper “The Quality of Accruals and Earnings: The Role of Accrual Estimation Errors” (see the 2002 supplement to Accounting Review, Volume 77, pages 35-59, www.jstor.org) indicate that firms with lower accrual quality actually had more accruals in the financial statements that were unrelated to cash flows compared to firms with higher accrual quality. Therefore, their model measures accrual quality by measuring the change in working capital and cash flows. The formula for this model combines the change in working capital and cash flow from operations for the current year compared to the total of total assets for both the current and prior year:
(cy Operating Cash Flow + ∆ cy WC) / (cy TA + py TA)i
While the accrual quality calculations cannot determine if the accrual manipulation is intentional or just an error, the calculations are very effective in pointing to anomalies in the financial statements, especially when using a dual-axis chart comparing the accrual quality to net income. When comparing net income to these calculations it is important to remember that the movement between net income and the accrual quality should correspond. When net income and the accrual quality movements are in opposite directions, then the accrual quality possibly represents accruals without attainable cash flows. Equally important to remember is that the Dechow-Dichev model may indicate anomalies in one period based on information from prior year accruals so the analysis must cover a two-year period for greater accuracy.
The following chart representing the comparison of the accrual quality to net income for XYZ, Inc. shows this variation. In period five, net income increases while the Dechow-Dichev accrual quality decreases. In period six, net income decreases while the Dechow-Dichev accrual quality increases. Because net income increased in period five while the Dechow-Dichev accrual quality decreased, the possibility of financial statement manipulation occurring in period five seems likely. Additionally, the change in period five presents the greatest concern since income increases and the Dechow-Dichev Accrual Quality decreases suggesting the accruals are not likely to produce achievable cash flows for the company. This chart actually indicates anomalies in financial statement data for period four, five and six.
In the same paper, Dechow-Dichev described earnings as earnings after short-term accruals but before long-term accruals as another method of comparing net income and cash flows from accruals. The formula for calculating earnings is actually part of the accrual quality formula:
cy Operating Cash Flow + ∆ cy WC = Earningsii
Just like the accrual quality model, the best method to analyze the Dechow-Dichev earnings is to compare the earnings to net income using a dual axis chart. One should expect to see a comparative relationship in the movement between the Dechow-Dichev earnings and net income, just like in the accrual quality model. When movement occurs in opposite directions, it is possible that the accruals in the financial statements are not providing realizable cash flows. Once again, the following chart from XYZ representing a comparison between net income and the Dechow-Dichev earnings indicates possible manipulation in the same periods as the accrual quality comparison to net income. This model indicates that net income associated with accruals will not generate cash flow for the company.
Continuing the analysis of the accruals related to XYZ, Inc.’s financial information, the next model, Sloan’s Accruals tests for an implied cash component of earnings from accruals recorded in the financial statements. Sloan’s research, published in July 1996 titled “Do Stock Prices Fully Reflect Information in Accruals and Cash Flows about Future Earnings?” (See Accounting Review, Volume 71, No. 3, page 293.) addresses earnings performance and the reflection of a company’s stock price based on the accrual and cash components of earnings.
Sloan suggests that a company may manipulate earnings with no future cash flows without investors understanding the cash and accrual components. Therefore, the investors would not understand that the reported earnings may not generate future cash flows. His model is somewhat complex and requires multiple steps to calculate. Yet, it is one of the models used effectively for various periods other than the customary periods for financial statement presentation.
• Implied cash component = net income +/- changes in current net operating assets
• Current net operating assets = current operating assets – current operating liabilities
• Current operating assets = total current assets – cash and cash equivalents
• Current operating liabilities = total current liabilities – (short-term debt + current portion of long-term debt + income taxes payable)iii
The ideal method to analyze Sloan’s Accruals is to compare the implied cash component, the accrual component and net income. The accrual component of the model is the change in current net operating assets. A positive accrual component indicates that accruals have increased net income while a negative accrual component indicates that accruals have decreased income. Therefore, a higher accrual component compared to net income and the implied cash component is a “red flag” suggesting that there are accruals in the financial statements that are not going to generate future cash flows.
The following table represents the accrual testing for XYZ, Inc. using the Sloan’s Accrual model. Based on the relationship of the accrual component to both the implied cash component and net income, periods four and six represent accruals recorded in the financial statement that will not generate future cash flows. Thus, the financial statements for these periods are subject to earnings manipulation.
The table also indicates that the accrual component has increased net income for both periods since the accrual component is positive, but without the underlying appearance of generating future cash flows for XYZ, Inc. since the implied cash component is negative.
By comparing the accrual testing for both the Dechow-Dichev model and the Sloan’s Accruals model, period six in both models indicates anomalies in the financial statement data. Two separate methods of accrual testing have identified the possibility of earnings management or “income-smoothing” in the financial statements of XYZ, Inc. for period six. However, why do the models differ about periods four and five?
On the surface, the two different models differ in analysis for both periods four and five until factoring in the movement trends associated with the Dechow-Dichev model’s formulas and the possibility that the anomaly found in one period may result from the prior period accrual activity. By using the Sloan’s Accruals model, the question posed in the Dechow-Dichev model concerning anomalies in both periods four and five that possibly represent earnings management is reduced to period four because Sloan’s model indicates the issues in period five from the Dechow-Dichev models relate to anomalies from period four that may suggest possible earnings management. From an investigative perspective though, using both models allow one to determine the exact period for the anomalies found in the Dechow-Dichev model and reduce the possibility of misinterpretation of its results.
The last model used to test accruals is Jones Nondiscretionary Accruals. This model is based on a study titled “Earnings Management During Import Relief Investigations” by Dr. Jennifer Jones in 1991. Her model determines to what extent discretionary accruals manipulate earnings. In order to understand the model, one must first understand the definition of a “discretionary” accrual.
Simply defined, a discretionary accrual is an expense recorded in the financial statements that is not required or mandatory, such as accruing expenses for management’s bonuses, compared to compulsory accruals such as accrued payroll where the expenses have been incurred but not paid at the end of the period. Sometimes accounts such as inventory reserves, allowances for bad debts and warranty reserves are considered discretionary accruals simply because these are subjective numbers based on management’s estimates and provide opportunity for earnings manipulation as well.
By measuring those compulsory or nondiscretionary accruals, Jones determined that one is able to measure discretionary accruals because, over time, the nondiscretionary accruals will equal zero. This model is especially effective for determining the possibility of “income-smoothing” or leveling out income from year to year in order for a company to attract investors.
The calculation for measuring nondiscretionary accruals using the Jones Nondiscretionary Accruals model is rather simple compared to the more complex calculations required in the Sloan’s Accruals model. The result of the formula noted below is a calculation of nondiscretionary accruals as a percentage of total assets.
(1 / TA py) + ((Rev cy – Rev py) / TA cy) + (PPE cy / TA py)iv
The nondiscretionary accruals on XYZ’s financial statements are similar to what one would find on any set of financial statements: accounts receivable, accounts payable, inventory, and accrued payroll. Moreover, the financial statements include the nondiscretionary accruals inventory reserves, allowance for bad debts and warranty reserves. The following table represents the results of Jones Nondiscretionary Accruals measurements.
The table clearly identifies that in period six that as the nondiscretionary accruals decrease the discretionary accruals increase. For XYZ this presented a betraying sign of earnings management through its inventory reserves. Furthermore, additional investigative work found evidence of inventory theft in addition to management’s “very liberal” calculations related to its inventory reserve account. Upon the discovery of both the use of unconventional methods to calculate the inventory reserves and the inventory theft that was corrected, the measurements of the nondiscretionary accruals increased in period four and similar to the calculation prior to the deception.
Based on the accrual analytical procedures performed on the financial statements for XYZ, Inc. periods four and six show anomalies from Sloan’s model and from the Dechow-Dichev model. When factoring in the results from Jones Nondiscretionary accruals, all three analytical procedures indicate the financial statements for period six imply accrual misrepresentation suggesting the possibility of “income-smoothing” or intentional earnings management.
The three techniques discussed in this article may be used for all types of entities and not be confined to a specific class. They provide simple methods to analyze accruals in financial statements and to measure their impact on the financial statements which are often overlooked in more traditional analytical techniques. Not only do they detect the possibility of fraudulent transactions, they also provide a means for measuring management’s use of “income-smoothing” in its financial statement presentations.
Pam Mantone, CPA, CFF, CFE, FCPA, CITP, CGMA, MAFF is a Director at Elliott Davis Decosimo in Chattanooga, Tennessee. Ms. Mantone specializes in litigation support services with emphasis on forensic accounting and fraud examinations. She has performed forensic and fraud auditing services for organizations, including the gathering of forensic evidence and testifying to findings. She also provides consulting services regarding implementation of fraud prevention and fraud detection internal control systems. Her experience includes conducting and supervising audits of local banks, credit unions, local not-for-profit organizations and HUD audits. She manages and performs external and internal audits of financial institutions. Ms. Mantone is an accomplished author. Her book, Using Analytics to Detect Possible Fraud – Tools and Techniques, was published in 2013 and provides a common source of analytical techniques used in forensic accounting investigations. It is also used as a college textbook.
iPatricia M. Dechow and Illia D. Dichev, “The Quality of Accruals and Earnings: The Role of Accrual Estimation Errors,” Accounting Review, Volume 77, Supplement: Quality of Earnings Conference (2002), page 15.
iiIbid, page 15.
iiiRichard D. Sloan, “Do Stock Prices Fully Reflect Information in Accruals and Cash Flows about Future Earnings?,” Accounting Review, Volume 71, No. 3, July 1996, page 293.
ivJennifer J. Jones, “Earnings Management During Import Relief Investigations”, Journal of Accounting Research, Volume 29, No. 2 (Autumn 1991), page 211.