The Department's auditor removed offsetting credit memos from my sales tax audit sample without a signed sampling agreement -- does that make the sample inflated and invalid?

Plain-English summary

A manufacturer and retailer of automotive tires, selling through national/regional retailers, its own stores, and independent dealers, was audited for sales and use tax on disposal fees, diagnostic fees, assets, and national account sales. Because of the sheer volume of transactions, the Department's auditor used the Invoice Capture Tool (ICT) sampling method -- a stratified statistical sampling approach -- to review compliance on the company's national account sales specifically. The company disputed the resulting assessment on those sales, arguing the sample was inflated and its error rate overstated.

Here's how national account sales worked: when a national account customer ordered, the company found the product at a dealer near the customer, pulled it from that dealer's inventory, and issued the dealer a credit memo reimbursing the dealer's initial (exempt, for-resale) purchase. The company argued that including BOTH the original sale (taxable) AND leaving out the offsetting credit memo (tied to the dealer's exempt purchase) inflated the total sales counted in the sample, overstating the error rate.

Sampling, generally, works by testing a representative slice of transactions, determining an error factor, and extrapolating that factor across the whole audit period -- a well-established, widely used audit technique that avoids the burden of a full transaction-by-transaction review. Under the Department's ICT sampling policy (previously published in P.D. 13-125), an auditor first removes any OFFSETTING positive and negative transactions (i.e., a sale and its matching credit that cancel each other out) from the sample population entirely; any remaining, unmatched negative transactions (credits without an offsetting sale) are then simply left OUT of the sample used to calculate the error factor. That's exactly what happened here -- and the auditor had explained this procedure to the company during the audit and given it a chance to supply invoices that could be matched against the remaining, unmatched credit memos. The company didn't provide such invoices during the audit, and still hadn't provided them on appeal.

The company cited three other rulings (P.D. 93-54, P.D. 95-47, and P.D. 10-179) to argue credit memos should always be counted, but the Department found those cases weren't analogous to the specific ICT-sampling procedure at issue here -- P.D. 13-125 was the directly applicable precedent. Finally, the company noted no signed sampling agreement existed between it and the auditor; Virginia confirmed that while having a signed agreement in place is a DESIRABLE practice under the Department's Field Audit Guidelines, the accuracy and validity of a sampling methodology doesn't depend on whether one was signed -- what matters is that the auditor properly explained and applied the procedure, which the record showed occurred here. The assessment was upheld as issued.

What this means for you

Businesses undergoing an ICT (Invoice Capture Tool) statistical sampling audit

If your sales involve offsetting credit memos (like a dealer-reimbursement or drop-ship model), understand that Virginia's established policy removes MATCHED offsetting pairs from the sample entirely, but excludes any REMAINING, unmatched credits from the error-rate calculation unless you supply invoices to match them. If you believe credits are being wrongly excluded, provide the matching invoices during the audit itself -- doing so only at the appeal stage, without documentation, generally won't succeed.

Businesses questioning a sample's validity because no signed sampling agreement was used

A signed sampling agreement is a best practice the Department recommends, but its absence doesn't invalidate an otherwise properly explained and applied sampling methodology. Don't rely on the lack of a signed agreement alone as a basis to challenge an audit sample.

Businesses citing prior published rulings to challenge a sampling result

Make sure the ruling you're citing actually addresses the SAME sampling methodology or issue you're disputing. Here, three cited rulings about credit memos generally were found not comparable to the specific ICT statistical-sampling procedure actually at issue.

Common questions

Q: How does Virginia's ICT sampling policy handle credit memos and offsetting transactions?
A: Offsetting positive and negative transactions (a sale and its matching credit) are removed from the sample population entirely; any remaining, unmatched credits are then excluded from the sample used to calculate the error factor, unless the taxpayer provides invoices to match them.

Q: Is a signed sampling agreement required for an audit sample to be valid?
A: No. It's a recommended best practice under the Department's Field Audit Guidelines, but the validity of the sampling methodology doesn't depend on having a signed agreement in place.

Q: What should I do if I think unmatched credit memos are being wrongly excluded from my sample?
A: Provide the Department's auditor with invoices matching those credits DURING the audit, when you're given the opportunity -- waiting until the appeal stage without documentation is unlikely to change the outcome.

Citations and references

• P.D. 13-125 (7/3/2013) (established the ICT sampling policy applied here: offsetting transactions removed, unmatched credits excluded absent matching invoices)
• P.D. 93-54 (3/5/1993), P.D. 95-47 (3/20/1995), and P.D. 10-179 (8/16/2010) (prior rulings on credit memos generally, cited by the taxpayer but found not analogous to this ICT sampling issue)

Subject

Audit : Statistical Sampling : ICT audit

Source

• Landing page: Virginia Laws, Rules & Decisions
• Ruling: P.D. 21-134

Original ruling text

October 26, 2021

Re: § 58.1-1821 Application: Retail Sales and Use Tax

Dear *:

This is in response to your letter submitted on behalf of * (the “Taxpayer”) in which you seek correction of the retail sales and use tax assessment issued for the period January 2014 through December 2016. I apologize for the delay in responding to your appeal.

FACTS

The Taxpayer, a manufacturer and retailer of automotive tires, sold its products through national and regional retailers, Taxpayer-owned stores and a network of independent dealers. A sales and use tax audit was conducted by the Department’s auditor for the period at issue. Assessments were issued with respect to disposal fees, diagnostic fees, assets and national account sales. Due to the volume of transactions during the sample month, the auditor used the Invoice Capture Tool (ICT) sampling methodology to review the Taxpayer’s sales and use tax compliance with respect to the national account sales. Based on this review, and in accordance with Department policy, a sales tax assessment was issued to the Taxpayer. The Taxpayer maintains that as a result of the sample methodology used by the Department’s auditor the sample is inflated, only in regard to its national account sales, and that the corresponding error rate is incorrect. Specifically, the Taxpayer contends that the sampling methodology used in auditing the national account sales is erroneous because the auditor removed credits from the sample population.

DETERMINATION

Sampling is an audit technique of significant value that is widely used in both the public and private sectors for all types of audits where a detailed audit would not prove beneficial either to the auditor or the client. When sampling techniques are properly applied, the final results are usually within a narrow percentage range of the actual amount that would have been determined by a detailed audit. The purpose of the audit sample is to determine a factor for errors within a representative selected period. Once the error factor is determined, the factor is extrapolated over the entire audit period. The purpose of the projection is to account for likely similar transactions on which Virginia tax has not been paid. The ICT sampling methodology provides a more comprehensive approach to reviewing a taxpayer's financial records. Use of the ICT makes it easier to draw a representative sample from which a smaller number of items may be examined in each stratum than to sample the total population.

When a sale was made by the Taxpayer to a national account customer, the Taxpayer located the desired product at a dealer located near the customer. The product was removed from the dealer’s inventory, and the Taxpayer issued a credit memo to the dealer to reimburse the dealer for its initial purchase of the product. The Taxpayer maintains that by including both sales (one taxable and one exempt) and excluding the credit memos (associated with the exempt sales), the total sales included in the sample is inflated and the corresponding error rate is overstated. The Taxpayer indicates that it provided information to the Department’s auditor showing that each credit was related to a corresponding purchase by the dealer. In addition, the Taxpayer states that a signed sample agreement was not entered into by the Taxpayer and the Department’s auditor.

The stratified sample for the national account sales at issue was determined based upon Department policy regarding sampling in ICT audits, as provided in Public Document (P.D.) 13-125 (7/3/2013). The auditor reviewed invoices (positive transactions) and credit memos (negative transactions) in the sample population. The auditor removed all offsetting negative and positive transactions from the population. The remaining negative transactions were not included in the population sampled by the ICT that was used to calculate the error factor for the sample. It is my understanding that during the performance of the audit, this procedure was explained to the Taxpayer, and the Taxpayer was given the opportunity to provide invoices to the Department’s auditor to match with any remaining credit memos.

In P.D. 13-125, the discrepancy at issue the difference between the ICT sample and the original transaction totals provided by the dealer. The auditor removed offsetting negative and positive transactions in accordance with Department policy and the dealer was informed that addition negative transactions that were removed from the population could be offset against any corresponding positive transactions included in audit exceptions. Because, the dealer did not provide any offsetting positive transactions to the negative transactions included in the, the Department concluded the total exception values were correct.

As provided in P.D. 13-125, the Department’s auditor properly removed offsetting invoices and credit memos from the sample population, and properly did not include the remaining credit memos in the sample population. During the performance of the audit, the Taxpayer was given the opportunity to provide invoices to offset any remaining credits in the audit period. The Taxpayer did not provide such invoices during the audit and has likewise not provided invoices with the appeal to offset any of the remaining credit memos. In accordance with established Department policy, the sample methodology used in the audit at issue is proper and is not inflated, nor is the error factor overstated.

The Taxpayer references P.D. 93-54 (3/5/1993), P.D. 95-47 (3/20/1995), and P.D. 10-179 (8/16/2010) in the appeal and maintains that the rulings illustrate the importance of taking credit memos. While credit memos or credits were at issue in each of the public documents cited, and were used in the audits to determine the assessments issued to each of the taxpayers, the facts in each are not analogous to the facts in the appeal at issue. As stated above, the determination in P.D. 13-125 is applicable in this instance, and the sample methodology used in the audit is proper.

Lastly, a sampling agreement was not entered into by the Taxpayer and the Department’s auditor. The Department’s Field Audit Guidelines provide that having a signed agreement in place is desirable. The sampling agreement is a tool that may be used by the Department. However, the accuracy of the sampling methodology and the sample population is not based upon whether an agreement is utilized. During the performance of the audit, the Department’s auditor explained the sample methodology procedure to the Taxpayer. The fact that a sampling agreement was not in place has no bearing on the validity of the sampling methodology or this determination.

Based upon this determination, the assessment is correct as issued. An updated bill, with interest accrued to date, will be mailed shortly to the Taxpayer. No further interest will accrue provided the outstanding assessment is paid within 60 days from the date of this letter.

The public documents cited are available on-line at www.tax.virginia.gov in the Laws, Rules and Decisions section of the Department’s web site. If you have any questions about this response, you may contact * in the Department’s Office of Tax Policy, Appeals and Rulings, at ***.

Sincerely,

Craig M. Burns

Tax Commissioner

AR/2105.P

Related Documents

13-125