Sampling Strategies in Sales Tax Audits Selecting a Sampling Methodology and Negotiating With Auditors

Transcription

1 presents Sampling Strategies in Sales Tax Audits Selecting a Sampling Methodology and Negotiating With Auditors A Live 110-Minute Teleconference/Webinar with Interactive Q&A Today's panel features: Dr. Will Yancey CPA, Dallas Roger Pfaffenberger, Director of Audit Sampling Practice, Ryan, Dallas Kenneth Helms, Tax Planning Manager, United Technologies Corp., Farmington, Conn. Brent Fryar, Director of Audits, Wal-Mart Stores Inc., Bentonville, AR. Wednesday, December 23, 2009 The conference begins at: 1 pm Eastern 12 pm Central 11 am Mountain 10 am Pacific You can access the audio portion of the conference on the telephone or by using your computer's speakers. Please refer to the dial in/ log in instructions ed to registrations. CLICK ON EACH FILE IN THE LEFT HAND COLUMN TO SEE INDIVIDUAL PRESENTATIONS. If no column is present: click Bookmarks or Pages on the left side of the window. If no icons are present: Click View, select Navigational Panels, and chose either Bookmarks or Pages. If you need assistance or to register for the audio portion, please call Strafford customer service at ext. 10

2 For CLE purposes, please let us know how many people are listening at your location by closing the notification box and typing in the chat box your company name and the number of attendees. Then click the blue icon beside the box to send.

3 Sampling Strategies In Sales Tax Audits Webinar Dec. 23, 2009 Dr. Will Yancey, CPA Roger Pfaffenberger Ryan Kenneth Helms United Technologies Brent Fryar Wal-Mart Stores Inc.

4 Today s Program Sampling Fundamentals, slides 3 through 11 (Will Yancey) Good, Mediocre And Ugly Experiences With States On Sampling, slides 12 through 18 (Ken Helms) Preventing Difficulties In Sample Audits; Hot Topics And Trends In Audit Sampling, slides 19 through 42 (Roger Pfaffenberger) Wal-Mart/Sam s Club Experiences, slides 43 through 49 (Brent Fryar) 2

5 Sampling Fundamentals 3

6 Links On Tax Audit Sampling pdf Last updated in January / li 4

7 When Is Sampling Appropriate? When taxpayer s records are voluminous When the cost and time to retrieve and examine supporting documentation for all items are high When sampling is authorized by statute or policy When taxpayer and auditor have sufficient trust in the sampling method and in the available data 5

8 Sampling Risk Sampling risk is the risk that a randomly drawn sample does not represent the results that would be achieved by examining every item in the population Statistical evaluation allows quantification of the sampling risk Example: 95% confidence $100,000 ± $8,000 Non-statistical evaluation depends on subjective judgment 6

9 Non-Sampling Risk Non-sampling risk is all other risk that the results are not true, such as: Measurement error: Incorrect interpretation of law and facts Biased selection: Sample results are only for tax underpayments, but population includes both tax underpayments and overpayments 7

10 Statistical Vs. Non-Statistical Under International Standard on Auditing (ISA) 530, statistical sampling means any approach to sampling that has the following characteristics: Random selection of a sample, and Use of probability theory to evaluate sample results, including measurement of sampling risk A sampling approach that does not have characteristics (a) and (b) is considered non-statistical sampling 8

11 Sampling Methods In SUT Audits Non-statistical block sampling Example: Select three months out of 36 in an audit period Non-statistical stratified random sampling Stratify population by account type or dollar range, or both Randomly select sample items Do not report a statistical confidence interval on results Such as Texas Comptroller s approach Statistical stratified random sampling Stratify population by account type or dollar range, or both Randomly select sample items Report a statistical confidence interval on results 9

12 Sample Size Most states have a minimum sample size per sampled stratum. Most states set their minimum between 100 and 300 Example: Expect a 1% error rate Minimum 3 errors per stratum Sample size = = items x 1% error rate = 3 errors per stratum t Thus, if the occurrence rate is low, you need large sample sizes per stratum 10

13 Agreements And Disagreements Try to negotiate a sampling agreement with the field auditor and computer audit specialist before the sample is selected Field auditors often lack experience with sampling; taxpayers need to coach them If disagreements occur, appeal to state s tax audit methods specialist or lead computer audit specialist If you request a sample expansion or an alternative sampling method, be careful what you ask for. You might get it and be surprised with the results! 11

14 Good, Mediocre And Ugly Experiences With States On Sampling 12

15 The 10 Commandments Of Sampling Thou shalt know thy data Thou shalt know the jurisdiction s s sampling procedures and authority Thou shalt always review the jurisdiction s most recent case law on sampling challenges Thou shalt never sample all G/L codes Minimize i i projection population Thou shalt confer and confirm multiple times with IT, A/P and A/R on the availability and completeness of all invoices/receipts i i in the audit period 13

16 The 10 Commandments Of Sampling (Cont.) Thou shalt never provide documentation to an auditor without a documented sampling plan that entails how common issues are to be resolved, e.g. missing invoices, numbers of samples/detail strata, projection calculations (% of dollar error, average dollar error per transaction, etc.), pilot sample procedures, credits and refunds Thou shalt never sign a jurisdiction s sampling agreement unless absolutely necessary Thou shalt never provide documentation to an auditor without first reviewing the data internally Thou shalt always confirm an auditor s calculations Thou shalt call for help before assessment 14

17 The Good Experience: Florida Knowledge of data and controls We researched Florida s rules and regs on sampling procedures Restricted population to specific accounts Confirmation of documents Reviewed documentation ti prior to submission i to the auditor No sampling agreement signed 15

18 The Mediocre Experience: Illinois Knowledge of data and controls We researched Illinois rules and regs on sampling procedures Restricted population to specific accounts Included fixed assets in sample methodology Incomplete sampling plan Additional liability established due to inclusion of certain transactions Signed sampling plan 16

19 The Ugly Experience: Texas Lack of knowledge of data Missing documentation Assumption that sample was a block vs. stratified No sampling plan 17

20 The Ugly Experience: Texas (Cont.) No notification by state of stratified sample No review of documentation before providing to auditor Minimal restriction on G/L codes Call for help after assessment 18

21 Preventing Difficulties In Sample Audits; Hot Topics And Trends In Audit Sampling 19

22 Know Data Well Prior To Audit Prepare brief descriptions of systems that produce audit data Prepare brief descriptions of how systems treat various types of transactions Research changes in accounting system or in chart of accounts during the audit period Examine prior audits and refund claims What problems arose? Is there potential for overlap? 20

23 Know Data Well Prior To Audit (Cont.) Research potential problem areas Frequency and nature of transactions in holding or suspense accounts (e.g., GR/IR in SAP) Extent of electronic transactions (e.g., evaluated receipts settlements) Procurement card exposure due to missing documentation Existence and treatment of duplicate payments to vendors Extent of out-of-state transactions Availability of resale exemption certificates and direct pay permits 21

24 Know Data Well Prior To Audit (Cont.) Consider the extent to which use tax accruals can be tied to tax returns Determine what is necessary to get from a record in the data file to a physical invoice Describe any patterns of fhigh h or low error rates among types of transactions Goal is to minimize chance that auditor with discover surprises in the audit data file 22

25 Discuss Data And Sampling Issues With Auditor Process of downloading data for the audit Reconciliation of download to the general ledger (G/L) and to the tax returns Sampling unit (line item, invoice, i other) Reconciliation is essential (if population data file not reconciled, then old adage applies: garbage in, garbage out ) 23

26 Discuss Data And Sampling Issues With Auditor (Cont.) Typical development of audit population data file Tax director requests download from IT department IT department produces download/data extraction Tax director reconciles download to G/L and tax returns Tax director analyzes data to gain knowledge and understanding of data (e.g., data summaries by accounts, vendors, year/month, tax codes, jurisdictions, etc.) Tax director and auditor agree on carve-down of downloaded data file to the audit and sampling population 24

27 Carving Down Taxpayer File To Targeted Audit Population 25

28 Carving Down Taxpayer File To Targeted Audit Population (Cont.) AP or sales data file is subset of the general ledger The population file contains the transactions within the scope of the audit The sample file is a subset of the population file and contains the transactions to be audited The tax accrual data file typically overlaps G/L, AP, population, and sample files Key elements in defining the audit population Sl Selection of accounts of fit interestt Stratification of transactions into groups with similar error rates (e.g., assets, expenses, P-cards, utilities, pre- and post-law change periods) 26

29 Understand The Audit Data Life Cycle Get Data Reconcile Data Define Population Sample Design Results Post Analysis 27

30 Issues To Negotiate Before Audit Division of population into strata based on qualitative and quantitative variables Qualitative stratification (i.e., categories representing types of transactions, time periods, outlets, plants) Required by law (e.g., legal entity) To manage risk (e.g., group for items likely to have missing documentation such as P-card transactions) To partition error rates (e.g., assets vs. expenses) 28

31 Issues To Negotiate Before Audit (Cont.) Division of population into strata based on qualitative and quantitative variables Quantitative stratification (i.e., numerical dollar ranges within groups) Lower exclusion threshold (don t audit items with absolute values less than exclusion threshold, e.g. $25) Upper detail threshold (take a complete census, or detail all items with absolute values greater than detail threshold (e.g., $100,000) Sample strata two but no more than seven mutually exclusive and exhaustive dollar ranges, from exclusion threshold to detail threshold 29

32 Issues To Negotiate Before Audit (Cont.) Determination of controlling date field (invoice date, posting date, etc.) Treatment of negatively valued transactions Treatment of missing records in audit Sampling method selected for the audit Statistical issues Block sample or stratified random sample Method of determining numerical strata breaks for stratified samples Sample sizes Detail thresholds Exclusion thresholds Tentative timetable for audit 30

33 Review Proposed Sampling Plan, Sampling Procedures Ensure that the sampling plan used to draw the sample is the plan that was originally agreed to Confirm agreements on the treatment of negatively-valued transactions and transactions with missing documentation Confirm the definitions, descriptions, and specifications of the population and the sample When auditing work is completed, check the accuracy of the estimation/projection methods and results Once the sample is pulled, check that the sample is random and representative of the population 31

34 Auditor Treatment Do: Appoint one contact person to whom the auditor may direct questions and requests and who is responsible for following up with the auditor Tie amounts reported in sales and use tax returns with G/L and to appropriate taxable or tax collection accounts. Prepare explanations for amounts that do not tie Provide auditor with requested documentation ti (e.g., G/L, tax returns, exemption certifications) Provide auditor with isolated and quiet work area with access to copy machine Meet with auditor to discuss accounting process and to select accounts of interest Jointly develop sampling gplan for audit with the auditor 32

35 Auditor Treatment (Cont.) Don t: Allow auditor unrestricted access to facility Allow auditor to have access to anyone other than appointed contact person Allow auditor to have access to internal computer systems or internal accounting databases Provide auditor with unabridged downloads of entire system files Ignore auditor s reasonable requests for information (e.g., tax returns, data files pertinent t to selected accounts of interest, t reconciliation reports) 33

36 Hot Topics And Trends: Missing Documentation Treatment 100% error in absence of documentation Use other evidence (e.g., invoices from same vendor for same or similar items showing tax paid) Use of replacement titem Georgia Ask Don Bloom to increase the sample size with the additional items used as spares (i.e., replacement items) if necessary and justified. Important note: You must ask Don to increase the sample size before he draws the sample Texas Added statement to Texas Sampling Manual (revised September 2008) to the effect that alternate information (e.g., invoices from same vendor showing tax paid) may be used to address issue of missing invoices 34

37 Hot Topics And Trends: Treatment Of Negatively-Valued Transactions Treatment Carve out of population and place in separate group Retain in population and stratify on absolute value Trend Place negatively-valued l ditems in separate group 35

38 Hot Topics And Trends: Minimum Number Of Errors To Project From Sample To Population Treatment Most states do not have a minimum number of errors to project California: At least 3 errors per stratum to project Texas: At least 2 errors per stratum to project Trend Use of minimum number of errors per stratum to project (at least 2) 36

39 Hot Topics And Trends: Overlapping Audits Treatment Varies by state Texas: When purchaser is audited, purchaser must show that exact same items was assessed in seller s audit Statistical view If item is in vendor/seller e population o (whether e in sample or not; and, if in sample, whether assessed or not), it should not be assessed in seller/purchaser audit 37

40 Hot Topics And Trends: Refunds Based On Taxpayer Samples Treatment t Varies by state; about 50% of the states allow refunds to be based on samples, and remaining states require refund claims to be based on detailed transactions Trend Modest trend in direction of more states allowing refunds to be based on samples, encouraging taxpayers to agree to enlarge the scope of the refund review to a full audit Trend States that allow refunds to be based on samples (e.g., Washington, Iowa) increasingly want to participate or require participation in the development of the sampling plan and prefer to the pull the samples 38

41 Hot Topics And Trends: Refunds Treatment Based On Auditor Samples Varies by state; many states allow overpayment to be projected within the audit sample but some do not (e.g., Georgia does not allow a net refund for sampled populations) Comment Goal of audit sample should be to estimate the correct amount of tax due which should result in one of three outcomes: (1) net underpayment (assessment), (2) net overpayment, (3) no tax due (neither e net underpayment e nor net overpayment) e If the audit results in a net overpayment, then the fair and equitable action is to refund the overpayment amount to the taxpayer 39

42 Tax Only Items in the Sample Treatment California: Treat as non-error item in sample Texas: Under certain circumstances, divide tax only amount by tax rate and add item to appropriate sample stratum, thereby creating a new sampled item Statistical view California has it right; treat the item as a non-error in the sample Comment The tax-only items should be excluded from the sampling population, if at all possible 40

43 Sampling Agreements Treatment t About half of the state have some form of a sampling agreement; some are binding (e.g., Ohio) and some are not (e.g., California) Having a written agreement that covers the key audit issues is desirable (e.g., California BOE Form 472), but binding agreements are unnecessary and dangerous (e.g., Ohio courts have limited taxpayer s ability to challenge audit if the sampling agreement has been signed) Trend Increased interest in having some form of formal documentation that the taxpayer and the auditor have met and discussed key issues and how they will be treated in the audit 41

44 Information Sources Multistate Tax Commission Sampling Manual; Internet: Federation of Tax Administrators Sampling for Sales and Use Tax Compliance white paper (December 2002); Summary of State Sampling Practices - Update on states audit sampling policies and procedures (January 2004). Internet: California Board of Equalization Chapter 13 Statistical Sampling in Audit Manual. Internet: Texas Comptroller of Public Accounts Sampling Manual (September 2008). Internet: e 42

45 Wal-Mart/Sam s Club Experiences 43

46 Regardless Of Type Of Audit, Know Your Company What does your company do (all lines of business)? How does your company accomplish these business activities? What are your company s goals? How does your business come across in an audit? 44

47 How Will Your Company Approach Determine the scope of the test The Audit? Identify likely specific areas of interest in the audit Know which locations have higher percentages of the variety of sales tested by the auditor Know how any information (business) has changed during the audit period Figure out if there has been any change of taxability of items during the audit 45

48 Narrowing Down Populations Knowing your business will help you determine what's important to be reviewed in the sample What sales you want to include in the audit process Unique opportunities within the process that you want to carve out and handle separately Eliminate areas that are not relevant or are immaterial to the process 46

49 Dealing With Homogenous Sales Gather the different types of sales into appropriate buckets Determine how to approach them 47

50 Ensuring That Adequate Records Are Available Know your business Negotiate and agree to replace certain invoices/sales with similar invoices/sales from the same supplier or customer Select a sample with more invoices than required, then use the excess to replace unavailable records 48

51 Conclusion Know your company s business before the state arrives Work hard to ensure that the auditor and you agree up front on how to handle different items, as they will inevitably show up in the process Work with the state to ensure that both parties both understand and agree to the process before you start 49