Exercise based on AICPA Guide- Audit Sampling- Chapter 5 Nonstatistical Sampling Case Study 5.01 This chapter provides a case study illustrating the design and use of a nonstatistical sample. 5.02 Sarah Jones of Jones & Co., CPAs, designed a nonstatistical sample to test the existence and gross valuation of the December 31, 20XX, accounts-receivable balance of Short Circuit Inc., a privately owned electrical supply company that is a continuing client of Jones & Co. For the year ended December 31, 20XX, Short Circuit had sales of approximately $25 million. As of December 31, there were 905 accounts receivable, with debit balances aggregating $4.25 million. These balances ranged from $10 to $140,000. There were also 40 credit balances aggregating $5,000.fn 1 5.03 In planning her audit, Sarah Jones updated her understanding of the client and its environment, including its internal control. She also understood that the entitys revenue recognition policy was to recognize revenue upon shipment. She also understood that cash sales are prohibited, and the entry for all sales transactions involves a debit to accounts receivable. In addition, all cash receipts are through the banks lock box, and there are no credits to income in the cash receipts journal. The only general journal entries affecting receivables and revenue involve minor write-offs of bad debts and setting up an allowance for doubtful accounts at the end of each quarter. All of the preceding were true in prior audits, and inquiry of client management indicates no changes from prior periods. 5.04 Jones made the following judgments in planning her procedures for revenue and receivables: Because this is not a first audit and because of some past errors in accounts receivables, her assessment of the risks of material misstatement in receivables did not support an assessment much below high for the assertions of existence and gross valuation of accounts receivable. Fraud risk related to revenue and receivables is low. There is little incentive to misstate revenue or receivables. The lock box system significantly reduces the risk of misappropriation of cash. The companys revenue recognition policy was appropriate in the circumstances. There was minimal risk of channel stuffing or other revenue recognition issues. The confirmation process would test existence and gross valuation of receivables. It would not provide much evidence about completeness, net valuation of receivables, presentation, and disclosure. Other procedures will be performed on the account that will address these assertions and provide some evidence on existence and valuation. The confirmation process would provide some evidence of the occurrence and gross valuation of sales transactions. This was because if receivables did not exist, the sales transaction did not occur. It also would provide some evidence about receivables cutoff because customers would report items included in receivables that were not shipped by year-end. The confirmation process would not provide evidence of completeness of revenue. 5.05 Sarah Jones made the following judgments in designing the confirmation sample: Her limited tests of controls supported an assessed level of risk of material misstatement (inherent and control risk) of high for the assertions of existence and gross valuation of accounts receivable. The preliminary assessment of overall materiality is $200,000. Tolerable misstatement for this test was set at $150,000, which is 75 percent of the materiality for the financial statements as a whole and less than the performance materiality set for the overall engagement. This judgment was based on the fact that most other accounts could be estimated to a significant precision and that the client would adjust for factual misstatements and follow up appropriately on projected misstatement issues, and this account had a higher expected error rate than other account balances.fn 2 She expects a possible $15,000 misstatement in accounts receivable, which is a realistic, or if anything, a somewhat conservative estimate based on the results of prior years testing. The credit balances in accounts receivable would be tested separately. The balance for each selected customer would be confirmed. As a percent of account receivables what does the $15,000 represent as %? What term from the power point presentation is most closely defined by this number (the%)? 5.06 She planned to project the sample result using a ratio method. What are some reasons for using the ratio method in relation the sampling strategy (stratification)? 5.07 The following is some additional information: The population contained 5 balances of more than $50,000, which totaled $500,000. Jones decided to examine these 5 balances 100 percent and exclude them from the population to be sampled. The population also contained 900 other debit balances, which totaled $3.75 million. Through substantive analytical procedures and cut-off tests, Jones obtained some assurance that all shipments were billed and that receivables were complete. What assertions of accounts receivable did Jones obtain evidence for through the analytic and cutoff tests? Determining the Sample Size 5.08 Considering the following factors, Jones determined the sample size: a. Variation in the population. Jones separated the population into 2 groups based on the recorded amounts of the items constituting the population. The first group consisted of 250 balances equal to or greater than $5,000 (total recorded amount of $2.5 million), and the second group consisted of the remaining balances that were less than $5,000 (total recorded amount of $1.25 million).fn 3 A computer program designed to interrogate populations electronically and select samples was used to efficiently perform this procedure. What is the term for breaking up the population the way Jones did? If the computer program was not available and Jones wanted to select the sample randomly, what are three techniques/tools Jones could have used accomplished this? b. Risk that a material misstatement does not exist, when it does. Referring to table 4-5, Illustrative Sample Sizes, Jones decided to use a 5 percent risk of incorrect acceptance (that is, desired confidence level of 95 percent). She based her decision on an assessed level of risk of material misstatement of high, limited assurance from substantive analytical procedures, and because she did not plan any other significant detailed substantive or control tests to achieve the same objectives.fn 4 c. Tolerable and expected misstatement. As indicated previously, the amount of tolerable misstatement for this sample was determined to be $150,000 and the amount of expected misstatement is estimated to be $15,000. 5.09 Jones then determined the appropriate sample size of 92 by referring to table 4-5 (5 percent risk that a sample will lead the auditor to conclude that a misstatement does not exist, when it does [risk of incorrect acceptance], 4 percent tolerable misstatement as a percent of the population, and 10 percent expected misstatement as a percent of tolerable misstatement). What would be the sample size if Jones used a 10 percent risk of incorrect acceptance, 4 percent tolerable misstatement as a percent of the population, and 20 percent expected misstatement as a percent of tolerable misstatement? 5.10 Jones considered the efficiency of other strategy alternatives to reduce the sample size and concluded the confirmation procedures would be the most efficient and effective approach. She considered performing further tests of controls; performing additional detailed, targeted analytical procedures; performing a test of sales transactions (a related financial statement area) that would also provide assurance on assertions relevant to this test; or increasing the number of items that are substantively tested 100 percent by lowering the threshold for selecting items for 100 percent testing below $50,000. However, she concluded that it would be more efficient to confirm 92 items. Per above The population contained 5 balances of more than $50,000, which totaled $500,000. What if items above $40,000 were 30 items and totaled $1.7 million. Would this change Joness decision? Why or why not? 5.11 She also decided to allocate the sample between the 2 groups in a way that was approximately proportional to the recorded amounts of the accounts in the groups. Accordingly, Jones selected on a haphazard basis 62 of the 92 customer balances from the first group or stratum (balances with recorded amounts equal to or greater than $5,000) and the remaining 30 customer balances from the second group or stratum (balances with recorded amounts under $5,000). Does the breakout of the sample as 62 and 30 make sense? Why or why not? Evaluating the Sample Results 5.12 Jones mailed confirmation requests to each of the 92 customers whose balances had been selected and to each of the 5 customers selected in the 100 percent examination group. Of the 97 confirmation requests, 82 were completed and returned to her. She was able to obtain sufficient assurance through alternative procedures that the 15 customer balances that were not confirmed were bona fide receivables and were not misstated. Of the 82 responses, 4 customers indicated that their balances were overstated.fn 5 Jones investigated these balances further and concluded that they were, indeed, partially misstated. She determined that the misstatements resulted from ordinary mistakes (for example, shipping charge variations, a misapplication of discount agreements, and credits) in the accounting process. The sample was summarized as shown in table 5-1, 100 Percent Examination and Sample Testing Summary. Table 5-1 100 Percent Examination and Sample Testing Summary Group Recorded Amount Recorded Amount of Items Tested Audited Amount of Items Tested Amount of Overstatement 100% examination $500,000 $500,000 $499,000 $1,000 Over $5,000 2,500,000 739,000 738,700 300 Under $5,000 1,250,000 62,500 62,350 150 $4,250,000 $1,301,500 $1,300,050 $1,450 5.13 Jones observed that the sample included 30 percent of the dollar amount of the over $5,000 group and 24 percent of the items included in that group. She also observed that the sample comprised 5 percent of the dollar amount of the under $5,000 group, and about 5 percent of the items included in that group. On the basis of the preceding computations, she considered the methods of projecting sample results described in this guide.fn 6 She considered the misstatements found and confirmed her previous judgment that the amount of misstatement in the population was more likely to correlate to the total dollar amount of items in the stratum or population than to the number of items in the stratum or population. As a percent how much of the total population was tested by Jones? What if the result was less than 5%, would Jones have to perform (a) no more work, (b) more work? Why? Thus, Jones decided to project misstatements based on the rate of misstatement in each group (stratum). Then Jones separately projected the rate of misstatement found in each groups sample to the total dollars from that group. For the over $5,000 group, she projected the sample results for that group to the population by multiplying the misstatement rate observed in the sample by the recorded amount for that group. She calculated the projected misstatement to be approximately $1,015 ($2,500,000 ($300 $739,000)). Similarly, Jones calculated a projected misstatement for the group under $5,000 to be approximately $3,000 ($1,250,000 ($150 $62,500)). Therefore, the total factual and projected misstatement from the items tested 100 percent and items sampled was $5,015 ($1,000 + $1,015 + $3,000). Management of Short Circuit Inc. agreed to correct the factual misstatements of $1,450, resulting in a remaining projected misstatement of $3,565. Why didnt Jones project the $1,000 difference/overstatement? What if management disagreed and did not correct for the misstatement? Is this a material misstatement? Should Jones continue the audit? 5.14 Jones compared the total factual and projected misstatement from the items tested 100 percent and items sampled of $5,015 with her $15,000 expectation of misstatement of accounts receivable and concluded that the sample results met the desired test objective. She also compared the remaining unadjusted misstatement ($3,565) with the $150,000 tolerable misstatement and determined that there was a small risk that this account could be misstated by more than the tolerable misstatement (of $150,000). In other words, there was an ample "cushion" between the tolerable misstatement and the remaining projected misstatement amounts to be able to conclude there is a low risk of material misstatement in the account. Jones investigated the nature and cause of the misstatements and determined that, as they resulted from explainable minor clerical error, they were not indicative of additional audit risk or a significant deficiency or material weakness in controls. AU-C section 265, Communicating Internal Control Related Matters Identified in an Audit (AICPA, Professional Standards), provides further guidance on evaluating the severity of control deficiencies identified in the audit. What if the projected differences were $2,000, $7,000 and $8,000 respectively for each category sampled? Would Jones reach a different conclusion? What if the projected differences were $14,999, $0 and $0 respectively for each category sampled? Would Jones reach a different conclusion? What if upon investigation it was determined the $5,015 projected error (total for all 3 categories) was the result of fraud. Would Jones be able to reach the same conclusions above? Would any additional procedures need to be performed? (if so what)? 5.15 Jones concluded that the sample results supported the recorded amount of the accounts-receivable balance; however, she did aggregate the remaining projected misstatement from the sample results with other factual and projected misstatements to evaluate whether the financial statements as a whole might have been materially misstated. Her evaluation of the potential material misstatement of the financial statements as a whole included considering qualitative factors, for example, trends and account relationships. (What are examples of qualitative factors that Jones might have seen per the information above?) What qualitative factors would indicate the account might be materially misstated? ________________________________________________ fn 1 The net population consisted of 945 balances with a total recorded value of $4,245,000. fn 2 Had all of these relevant factors been considered when setting performance materiality, then performance materiality and tolerable misstatement may have been set to the same amount. fn 3 Had the population not been stratified, the sample size would have been increased (see chapter 4, Nonstatistical and Statistical Audit Sampling for Substantive Tests of Details) due to the variability of the items in the population. fn 4 Had control tests been performed and supported effective controls, an acceptable risk higher than 5 percent (lower desired assurance) would likely have been appropriate. The extent of reduced assurance for this substantive test would be responsive to the extent of controls testing and the control test results. The design and performance of effective analytical procedures, for example by meaningful subclasses of receivables, can also reduce the extent of substantive detailed testing. fn 5 Three were in the sample group and one was in the 100 percent tested group. fn 6 Had Jones selected the sample attempting to approximate a probability proportional to size selection, the monetary unit sampling point estimator described in chapter 6, Monetary Unit Sampling, might also be used. Table 4-5 Illustrative Sample Sizes Tolerable Misstatement as a Percentage of Population Risk of Incorrect Acceptance Ratio of Expected to Tolerable Misstatement 50% 30% 10% 8% 6% 5% 4% 3% 2% 1% 0.50% Expected Sum of Taints 5% 6 10 30 38 50 60 75 100 150 300 600 5% 0.10 8 13 37 46 62 74 92 123 184 368 736 0.37 5% 0.20 10 16 47 58 78 93 116 155 232 463 925 0.93 5% 0.30 12 20 60 75 100 120 150 200 300 600 1,199 1.80 5% 0.40 17 27 81 102 135 162 203 270 405 809 1,618 3.24 5% 0.50 24 39 116 145 193 231 289 385 577 1,154 2,308 5.77 10% 5 8 24 29 39 47 58 77 116 231 461 10% 0.20 7 12 35 43 57 69 86 114 171 341 682 0.69 10% 0.30 9 15 44 55 73 87 109 145 217 433 866 1.30 10% 0.40 12 20 58 72 96 115 143 191 286 572 1,144 2.29 10% 0.50 16 27 80 100 134 160 200 267 400 799 1,597 4.00 15% 4 7 19 24 32 38 48 64 95 190 380 15% 0.20 6 10 28 35 46 55 69 91 137 273 545 0.55 15% 0.30 7 12 35 43 57 69 86 114 171 341 681 1.03 15% 0.40 9 15 45 56 74 89 111 148 221 442 883 1.77 15% 0.50 13 21 61 76 101 121 151 202 302 604 1,208 3.02 20% 4 6 17 21 27 33 41 54 81 161 322 20% 0.20 5 8 23 29 38 46 57 76 113 226 451 0.46 20% 0.30 6 10 28 35 47 56 70 93 139 277 554 0.84 20% 0.40 8 12 36 45 59 71 89 118 177 354 707 1.42 20% 0.50 10 16 48 60 80 95 119 159 238 475 949 2.38 25% 3 5 14 18 24 28 35 47 70 139 278 25% 0.20 4 7 19 24 32 38 48 64 95 190 380 0.38 25% 0.30 5 8 23 29 39 46 58 77 115 230 460 0.69 25% 0.40 6 10 29 37 49 58 73 97 145 289 578 1.16 25% 0.50 8 13 38 48 64 76 95 127 190 380 760 1.90 30% 3 5 13 16 21 25 31 41 61 121 241 30% 0.20 4 6 17 21 27 33 41 54 81 162 323 0.33 30% 0.40 5 8 24 30 40 48 60 80 120 239 477 0.96 30% 0.60 9 15 43 54 71 85 107 142 213 425 850 2.55 35% 3 4 11 14 18 21 27 35 53 105 210 35% 0.20 3 5 14 18 23 28 35 46 69 138 276 0.28 35% 0.40 4 7 20 25 34 40 50 67 100 199 397 0.80 35% 0.60 7 12 34 43 57 68 85 113 169 338 676 2.03 50% 2 3 7 9 12 14 18 24 35 70 139 50% 0.20 2 3 9 11 15 18 22 29 44 87 173 0.18 50% 0.40 3 4 12 15 19 23 29 38 57 114 228 0.46 50% 0.60 4 6 17 22 29 34 43 57 85 170 340 1.02 4.69 Table 4-5 might also help the auditor understand the risk level implied by a given sample size. For example, the auditor might be designing a nonstatistical sampling application to test a population of 2,000 accounts receivable balances with a total recorded amount of $1 million. The auditor may have