6/11/2013. Learning Objectives. Points to Remember! USING ANALYTICS TO DETECT POSSIBLE FRAUD Overview of Techniques

Transcription

1 A Global Reach with a Local Perspective USING ANALYTICS TO DETECT POSSIBLE FRAUD Overview of Techniques Learning Objectives Performing various analytical indices to financial statements Using visual techniques not just the numbers Identify anomalies in trends that require further investigation for possible fraudulent transactions and/or errors in financial statements. Don t rely on judgment alone to identify anomalies in trends use empirical techniques such as chi-square tests, z- score calculations, t-tests - just to mention a few of the many statistical tests that may be performed. Points to Remember! Analytical procedures by themselves are not sufficient for prosecution but are tools that provide a road map for areas requiring further investigation. Choose analytical procedures that meet the needs of the engagement and apply results to benchmarks. Financial records are not always totally conclusive of all transactions so you must consider transactions occurring outside of the financial records. For example, a company may have off-book bank accounts whereby the bank accounts and the transactions within these accounts were not recorded in the financial records. More detailed evidence is essential for prosecution and multiple methods must be utilized in further investigations. 1

2 Preliminary Basic Analytical Tests Liquidity/Debt Ratios Vertical Analysis Profitability Ratios Ratios alone are seldom sufficient so forensic indices should be performed. Ratios were originally developed by lenders emphasizing collateral, capacity, liquidity, etc. Signs That Warrant Further Inspection Abnormal profitability when similar companies are not making profits within the same parameters Recurring negative cash flows from operations while reporting earnings and growth Abnormal growth in Days Sales in Receivables compared to benchmarks of similar companies. (RMA, BizStats, etc) Abnormal increase in growth profit margin or the growth profit margin is in excess of industry standards Liquidity/Debt Ratios Working Capital Current Assets - Current Liabilities Working Capital Index Current year Working Capital - Prior Year Working Capital Current Ratio Current Assets/Current Liabilities 2

3 Liquidity/Debt Ratios Days Payable Outstanding 365 x (Average Accounts Payable/Cost of Sales) Days Sales Outstanding 365 x (Average Accounts Receivables/Sales) or 365 / Accounts Receivable Turnover (Sales/Average Accounts Receivables) Inventory Turnover COGS/((Beginning INV + Ending INV)/2)) Liquidity/Debt Ratios Cash Flow to Capital Expenses Cash Flow from Operations / Capital Expenses Cash Flow to Debt Cash Flow from Operations / Total Debt Obsolete Inventory Ratio Obsolete Inventory / Ending Inventory Net sales divided by (current assets current liabilities) Larger values considered desirable Significant increases compared to benchmarks could be symptom of revenue-related fraud 3

4 Gross Profit Sales Cost of Sales Profitability Ratios Sales Growth Index Current Year Sales / Prior Year Sales Gross Margin Sales Cost of Sales / Sales Return on Equity Net Income/((Beginning SE + Ending SE/2)) Profitability Ratios Gross Margin Index CY Gross Margin / PY Gross Margin Stock Sales Ending Inventory / Sales Return on Equity Net Income/((Beginning SE + Ending SE/2)) 4

5 Other Types of Analysis Analyses of Trends Over a Period of Time Comparing any of the analytical ratios over more than two periods the more the merrier! Process of analyzing trends and changes over more than two periods allows a review of financial data over a greater period of time for comparisons of anomalies that may build over time. More Advanced Analytical Forensic Indices 5

6 Beneish s M-Score 1 M= *DRSI *GMI *AQI *SGI *DEPI 0.172*SGAI *TATA -.327*LVGI. > suggests a higher probability of financial statement manipulation. Days Sales in Receivable Index (DRSI) general benchmark of 1 but varies based upon company characteristics - measures earning manipulation through receivables Gross Margin Index (GMI) - general benchmark of 1 but can vary as noted emphasizes even small amounts of change for further review Asset Quality Index (AQI) general benchmark of 1 but can vary as noted measures change in asset realization Sales Growth Index (SGI) general benchmark of 1 but can vary as noted measures rate of growth Beneish s M-Score 1 M= *DRSI *GMI *AQI *SGI *DEPI 0.172*SGAI *TATA -.327*LVGI. > suggests a higher probability of financial statement manipulation. Depreciation Index (DI or DEPI) general benchmark of 1 but can vary as noted shows potential of earnings manipulation through depreciation rate SG&A expenses (SGAI or SGAEI) general benchmark of 1 and fairly stable measures potential manipulation Leverage Index (LI or LVGI) general benchmark of 1 but can vary as noted measures debt loading Total Accruals to Total Assets (TATA) general benchmark of 0 and fairly stable measures potential earnings manipulation DSRI = (cy A/R/Sales) / (py A/R/ Sales) GMI = cy GM / py GM AQI = (1-(cy CA + cy Net FA)/cy TA)) / (1-py CA + py Net FA) / py TA)) SGI = cy Sales / py Sales DEPI = (py DE / py DE + py Net PPE) / (cy DE/cy DE + cy Net PPE) 6

7 SGAI = (Cy SG&A/Cy Sales) / (Py SG&A/Py Sales) LI = ((cy LTD + cy CL) / cy TA) / (py LTD + CL)/py TA) TATA = ((cy WC py WC)-(cy Cash py Cash) + (cy Income Tax payable py Income Tax payable) + (cy Current LTD py Current LTD) cy DE)) / cy TA cy current year py prior year A/R accounts receivable GM gross margin CA current assets FA fixed assets DE depreciation expense PPE property, plant and equipment SG&A sales and administrative expenses LTD long term debt TA total assets (CY Operating Cash Flow + the change in Working Capital for CY)/(CY Total Assets + Total Assets PY) Measurement combines the change in working capital and cash flows from operations for the current year compared to total assets. Simplest and most effective method is to compare with net income. 7

8 Jones Nondiscretionary Accruals 3 By measuring nondiscretionary accruals, you are indirectly measuring discretionary accruals which may indicate potential earnings manipulation. For example, as nondiscretionary accruals decrease, discretionary accruals increase. (1/PY Total Assets) + ((CY Revenue PY Revenue)/CY Total Assets) + (CY PPE/PY Total Assets) Lev-Thiagarajan 12 Signals 4 Score based upon 1 point each for positive, negative and neutral comparing each year to the preceding year. Each of the 12 factors are equally weighted. Negative signals may indicate a decrease in gross margins disproportionate to sales, disproportionate decreases in capital expenditures and R & D compared to industry standards, increases in S&A expenses disproportionate to sales, and unusual decreases in effective tax rates. Significant negative score in a given year warrants investigation. Inventory - % change in Inventory - % change in Sales. Disproportionate inventory increases considered a negative signal. (Index positive.) The Twelve Signals AR - % change in AR - % change in Sales. Disproportionate increases in AR considered a negative signal. (Index positive) Capital Expenditures - % change in industry benchmarks - % change in company s expenditures. Disproportionate decreases relative to benchmarks considered a negative signal. R & D Costs - % change in industry benchmarks - % change in company s expenditures. Disproportionate decreases relative to benchmarks considered a negative signal. S & A Expenses - % change in selling and administrative expenses - % change in sales. Disproportionate increases in S & A expenses considered a negative signal.( Index positive) Gross Margin - % change in gross margin - % change in sales. Decrease in gross margin relative to sales considered a negative signal. (Index negative.) 8

9 The Twelve Signals Allowance for Doubtful Accounts - % change in gross AR - % change in the Allowance. Positive values considered a negative signal. Effective tax rate - Tax Expense/Pre-tax Income. Unusual decrease (index is a negative number) considered a negative signal. Order backlog - % change in sales - % change in order backlog. A decrease in order backlog relative to sales is considered a negative signal. Labor costs - % change in sales per employee. Decreases in sales per employee (index number is positive) is considered a negative signal LIFO Earnings- LIFO cost of sales more closely approximates current cost compared to FIFO. Use of LIFO is considered a positive signal. Audit Opinion A qualified, disclaimed or adverse opinion is considered a negative signal. Includes nine variables from a company s financial statements. Originally used for evaluating a stock s financial strength. Does not require any market values so it can be applied to private company financial statements. Based upon a point system of 1 and zero Scores higher than 8-9 points suggest stronger stock and 0-2 points suggest weakest points The Nine Variables If CY net income is positive 1 point If CY operating cash flow is positive 1 point If CY operating cash flow exceeds net income 1 point If PY operating cash flow exceeds PY net income 1 point If ratio of Long-term debts to total assets decrease over PY 1 point If current ratio increases from PY 1 point If shares outstanding is no greater than PY (measure of potential dilution) 1 point If gross margin increases from PY 1 point If % increase in sales exceeds % increase in total assets 1 point 9

10 Sloan s Accruals Negative association between accounting accruals (the non-cash component of earnings) and subsequent stock returns Calculates the implied cash component of earnings by computing the changes in current net operating assets. Net operating assets defined as 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) A positive accrual component indicates accruals have increased net income and should be deducted from net income in order to obtain the implied cash component. A negative accrual component indicates that accruals have decreased net income and should be added to net income to obtain the implied cash component. Used in forensic analysis to focus on specific time periods for further detailed analysis. Other Types of Analysis Stratifications Monte Carlo Simulation Aging Regression analysis Digital Analysis / Benford s Law Stratification Monte Carlo Simulation Regression Analysis Aging Benford s Law 6 Segments population into groups of items exhibiting similar value amounts and sample each stratum separately Mathematical technique used to approximate the probability of certain outcomes by running multiple trial runs (simulations) using random variables Analyzes risks Other tools may be easier to use and provide similar results Used to predict a dependent variable Monitor the reasonableness of a value Assess the impact of dependent drivers Method sorting population by date for further examination of older transactions Digit and digit sequences in a data set follow a predictable pattern Human choices are not random but invented numbers that probably will not follow Benford s Law! Viewed graphically anomalies of the digit and digit sequence process will exhibit a different shape compared to the shape predicted by Benford s Law 10

11 The Z- Score Not to be confused with the Altman Z score Chebyshev s Theorem for Nonsymmetrical Distributions 75% of all observations fall within 2 standard deviations of the mean 89% of all observations are within 3 standard deviations of the mean 94% of all data are within 4 standard deviations of the mean The Z- Score (Data point-mean)/standard deviation Empirical Rule for symmetric distributions About 68% of all observations are within one standard deviation of the mean About 95% of all observations are within two standard deviations of the mean Almost all (more than 99%) of all observations are within three standard deviations of the mean Conversion of Z-Scores to Percentages Use the NORMSDIST formula (Excel 2007) - or NORMS.S.DIST formula (Excel 2010) 33 A/R z score (negatives not relevant when comparing to the Empirical Rule or Chebyshev s Theorem since direction applies for either side) NORMS.S.DIST(0.36, True) =.641 or 64.1% 11

12 Combining Z-Scores and the Empirical Rule 7 If the Z-score is less than -2 or more than 2 If the Z-score is less than -3 or more than 3 Occurrence about 5% of the time Occurrence less than 1% of the time 34 Unusual and possibly an outlier Very unusual and probably an outlier 35 These forensic indices are just a few of many that can be used in analyzing anomalies in financial data. Try other methods, including various types of statistical tools such as correlation analysis and chi-square tests, use of the net-worth method and use of the sources of funds methods just to name a few! 36 12

13 Quick Tips Develop and perfect your analytical and visual presentation skills Plan the engagement by learning to identify and then focus on questionable areas Don t rely on just your judgment Use multiple methods to prove your case Sources of Information 1 The Detection of Earnings Manipulation, Professor Messod D. Beneish, June, The Quality of Accruals and Earnings: The Role of accrual Estimation Errors, Professors Patricia M. Dechow and Ilia D. Dechev, July, Earnings Management During Import Relief Investigations, Jennifer J. Jones, JOURNAL OF ACCOUNTING RESEARCH, Vol. 29, No. 2, Autumn, Fundamental Information Analysis, Baruch Lev and Ramu Thiagarajan, JOURNAL OF ACCOUNTING RESEARCH, Vol. 31, No. 2, Autumn, Value Investing: The Use of Historical Financial Statement Information to Separate Winners from Losers, Joseph D. Piotroski, JOURNAL OF ACCOUNTING RESEARCH, Vol. 38, Supplement, January, The Use of Benford s Law as an Aid in Analytical Procedures, Mark J, Nigrini and Linda J. Mittermaier, AUDITING: A JOURNAL OF PRACTICE & THEORY, Vol. 16, No. 2, Fall, 1997 and I ve Got Your Number, Mark J. Nigrini, JOURNAL OF ACCOUNTANCY, May, Pelosi, Marilyn K. and Sandifer, Theresa M., Doing Statistics for Business with Excel, John Wiley and Sons,