Go to navigation Go to content
Toll-Free: (855) 743-0262
Phone: (201) 529-8024
Fazzio Law Offices
Call (201) 529-8024
Toll Free (855) 743-0262

Reasonable Methodologies in Sales and Use Tax Audits

NYS Sales and Use Tax Restaurant Audits

Restaurants – Reasonable and Not Arbitrary Audit Methodologies

The sales tax assessment can be based on a number of different methodologies, including a utility factor. Under New York Tax Law § 1138(a) the Commissioner of Taxation is authorized to use any information including estimation based on external indices. The one limit on sampling by the New York State Department of Taxation and Finance is that its methods must not be arbitrary. Grecian Square, Inc. v. State Tax Commission, 119 AD2d 948, 501 NYS2d 219, (New York Supreme Court Appellate Division - 3rd Department, 1986). In order to be “reasonable” and not “arbitrary” the audit method must be “reasonably calculated to determine the amount of tax actually due.” A reasonable calculation must take into account the attributes of the particular business in question. When that business is a restaurant, the auditor must consider the size of the restaurant, the number of employees, the number of seats, the seat turnover, and whether the restaurant is larger or smaller than the median or mean averages utilized.

Lack of Adequate Records Will Cause You to Lose Your Audit – Don’t Get Caught Unprepared!

Recently, the Division of Tax Appeals dealt with a sales tax case in which a pizzeria lacked adequate records, and a rent factor was used to estimate taxable sales (only works for rentals). Constantini v. State Tax Commission, DTA No. 820590, (New York Division of Tax Appeals, February 1, 2007). In Constantini, the taxpayer failed to present clear and convincing evidence to invalidate the auditor's methodology. Id. However, the finding that an index was appropriate in that case was based solely on the fact that there were no records produced by the taxpayer that the Auditor might have used to reasonably calculate sales; but, that case did not deal with whether the estimation method chosen was a reasonable one for calculating sales tax for a pizzeria of that size, number of seats, and number of employees. It was conceded that the averages would be applicable, and only the actual rental figures were contested.

The legal standard on review for a sales tax audit relying on external indices was set forth in Matter of AGDN, Inc. (New York Division of Tax Appeals, February 6, 1997), as follows:

a vendor . . . is required to maintain complete, adequate and accurate books and records regarding its sales tax liability and, upon request, to make the same available for audit by the Division (see, Tax Law §§ 1138[a]; 1135; 1142[5]; see, e.g., Matter of Mera Delicatessen, Tax Appeals Tribunal, November 2, 1989). Specifically, such records required to be maintained "shall include a true copy of each sales slip, invoice, receipt, statement or memorandum" (Tax Law § 1135). It is equally well established that where insufficient records are kept and it is not possible to conduct a complete audit, "the amount of tax due shall be determined by the commissioner of taxation and finance from such information as may be available. If necessary, the tax may be estimated on the basis of external indices . . ." (Tax Law § 1138[a]; see, Matter of Chartair, Inc. v. State Tax Commn., 65 AD2d 44, 411 NYS2d 41, 43). When estimating sales tax due, the Division need only adopt an audit method reasonably calculated to determine the amount of tax due (Matter of Grant Co. v. Joseph, 2 NY2d 196, 159 NYS2d 150, cert denied 355 US 869); exactness is not required (Matter of Meyer v. State Tax Commn., 61 AD2d 223, 402 NYS2d 74, lv denied 44 NY2d 645, 406 NYS2d 1025; Matter of Markowitz v. State Tax Commn., 54 AD2d 1023, 388 NYS2d 176, affd 44 NY2d 684, 405 NYS2d 454). The burden is then on the taxpayer to demonstrate, by clear and convincing evidence, that the audit method employed or the tax assessed was unreasonable (Matter of Meskouris Bros. v. Chu, 139 AD2d 813, 526 NYS2d 679; Matter of Surface Line Operators Fraternal Org. v. Tully, 85 AD2d 858, 446 NYS2d 451).

On appeal, a taxpayer needs to show by clear and convincing evidence that the tax assessed was unreasonable. Where formal books and records are not available, the commissioner may rely upon “such information as may be available.” Therefore, it is permissible, for instance, and logical, to rely on vendor invoices and to utilize a numerical calculation of actual sales based on raw material inputs. On the contrary, it is not logical to compare a small restaurant to an industry average without taking into account the “per seat” average of that group and comparing the small restaurant to other pizzeria’s with like size, volume, and staffing characteristics.


There is no question that under the law the external indices chosen must be comprised of data that is “applicable to petitioner’s business” (i.e. correlates to the size, square footage, and volume of sales) See 33 Virginia Place v. Comm’r, supra.


Mere Reliance on a Markup Percentage from an External Index Is Not Enough


In 33 Virginia Place v. Comm’r, DTA Nos. 821181, 821182, 821183, 821290, 821291 & 821859 (N.Y.S. Tax App. Trib, Mar. 31, 2011), the Division of Tax Appeals found that mere reliance on an index is not enough. The Judge found that the way the index was used in the audit was unreasonable in a number of ways:

§  The auditor failed to calculate median sales per seat (would have resulted in sales of $1 million less than reported).


§  The auditor did not consider the number of seat turns per day (experts testified that the high number of seat turns required to get the auditor’s figures was unreasonable).


§  The auditor’s found markup percentages of 281% for food and 285% for beverages, but never bothered to check whether such high markup percentages were commonplace in the industry and geographic area.


The Judge in 33 Virginia Placee v. Comm’r found that the Auditor must be familiar with the indices used and be able to explain how its use is “reasonably calculated to determine the amount of tax due” when applied to the subject business:

In summary, it is undisputed that Mother's did not maintain and, therefore, produce for audit, the books and records necessary to perform a detailed audit. Case law holds that, as a general proposition, any imprecision in the results of an audit arising by reason of a taxpayer's own failure to maintain adequate and accurate records of all of its sales as required by Tax Law § 1135(a)(1) must be borne by that taxpayer (Matter of Meyer v. State Tax Commn.; Matter of Markowitz v. State Tax Commn.).

What resulted from the two sales tax audits at issue in this matter is far more than imprecision, however. Both auditors chose to utilize data contained in a Restaurant Industry Operations Report by the National Restaurant Association and Deloitte & Touche LLP. The first auditor used the 2002 edition; the second auditor used the 2006/2007 edition. While prior Tax Appeals Tribunal decisions have sustained the use of this Report, it must be pointed out that in a number of these cases, the only data utilized was a rent factor. Obviously, the editors of the Report have provided sufficient warning as to its year-to-year reliability and its use for anything more than a management tool. . . . Since these auditors nevertheless chose the Report as a means by which to compute sales tax liability, they must be sufficiently familiar with the Report to be able to respond meaningfully to inquiries regarding its use (see Matter of Fokos Lounge) and, clearly, such familiarity was lacking herein.

The auditors could not explain why this particular report was chosen, and they did not sufficiently familiarize themselves with the contents so as to be certain that the chosen data was applicable to petitioner's business. Each auditor attempted to calculate a markup percentage which was applied to petitioner's purchases. Neither auditor bothered to check the reasonableness of their markup percentage by comparing the resulting audited taxable sales with other data which was set forth in the same section of the report that was utilized in computing the markup percentage.

Petitioner, while clearly negligent in its record keeping, was most diligent in analyzing the audit results in both sales tax audits. Its witnesses, most notably, David E. Gross and Brendan McCafferty, visited the business premises, reviewed the reports used by the auditors and performed independent analyses of the data contained in the reports which, in total, leads to the conclusion that the methods selected were not reasonably calculated to reflect tax due. By clear and convincing evidence, petitioner has shown that the results of both sales tax audits were erroneous and, clearly and unequivocally, that the audit methods employed were unreasonable.

In overturning the audit in 33 Virginia Place v. Comm’r, supra, the Division of Tax Appeals found that the number of seats in the restaurant was an important factor, and that an auditor should have “customized” the index data to the particulars of the restaurant size involved – which in this case would have resulted in a sales tax figure of $1,000,000 less than that arrived at by the auditor:

There is a distinction to be made between determining whether or not it was rational for the Division to use particular external indices, such as the 2002 and 2006/2007 Reports, as the basis for its audit and whether the audit methodology resulting from the use of these Reports was reasonably calculated to reflect taxes due. We find that in this case, it was rational for each auditor to rely on the Restaurant Industry Operations Report as the basis for calculating petitioners' sales and use tax liability, notwithstanding each auditor's inability to testify as to the quality of the data contained in the report.

That being said, however, the manner in which these indices were used resulted in audit methodologies that cannot be said to have been reasonably calculated to reflect petitioners' tax liability.

* * *

Had the Division used the median portion of Exhibit C-12 appropriately in calculating petitioners' taxable sales, it would have estimated petitioners' tax liability for the period by calculating the sales price of food and beverages sold "per seat," based on the number of seats in petitioners' restaurant. As pointed out by the Administrative Law Judge, this calculation, if performed, would have shown that audited sales were significantly lower than sales actually reported by petitioners, and resulted in no additional tax due by petitioners. In fact, even if the auditor had considered petitioners' business to be in the upper quartile of Exhibit C-12 instead of in the median range, a "per seat" analysis would still have shown petitioners' taxable sales to have been lower than what it had previously reported.

Rather, what the Division did was to extract numbers from Exhibit C-12 of the report and use those numbers in a manner for which they were not intended. Whatever information was used to compile the cost of beverages and food and the sales of beverages and food contained in Exhibit C-12 of the report was necessarily modified by the number of seats in the restaurants that contributed to their cost and sales figures.

Ignoring this component of Exhibit C-12, the Division used the Restaurant Report to create its own external index, which it identified as a "markup percentage." The term "markup percentage" is not contained in Exhibit C-12.

The Division is correct that its auditors are not required to explain the basis of the data underlying the external indices used, nor the method in which the data was formulated. We believe, however, that this does not give the Division carte blanche to simply extract convenient numbers from an index and use them in a manner for which they were never intended to be used. Having chosen to use the median portion of Exhibit C-12 of the 2002 Report, the Division should have used it for the purposes for which it was intended, and not randomly selected numbers from the exhibit that supported an increased tax liability for petitioners. While there is no question that petitioners' lack of records opened the door for the Division to use an external index to estimate petitioners' taxable sales, it must be pointed out that a lack of records does not equate to a presumption that taxable sales have been underreported.

While it is advisable to do so, auditors are not required by law to confirm the reasonableness of the results of their chosen methodology. However, the failure to do so here proved fatal to the Division's choice of audit methodology. The choices of exhibits were the Division's to make and, having made them, the Division is responsible for the outcome.

* * *

While the Division is not under an obligation to use an external index in such a way as to minimize petitioners' tax liability, as stated above, the index chosen must be "reasonably calculated to reflect the taxes due" (see, Matter of W.T. Grant Co. v. Joseph, supra). Had consideration been given to other exhibits in that Report, it would have appeared obvious that petitioners may not have underreported sales in any respect. With no articulated basis having been provided for using one exhibit over another, and given the divergent range of results depending on the choice of exhibits within the same external index, we find that, in this instance, the use of Exhibit C-16 of the Report by the auditor to calculate petitioners' taxable sales was not "reasonably calculated to reflect the taxes due" by petitioners.

The Experts in 33 Virginia Place determined that the seat turnover rate implied by the Auditor’s findings was unreasonable compared with industry standards. They found that the auditor failed to consider the sales per seat. The experts testified that the subject business did not have an employee-count near the average. The experts also testified that there is wide variability in sales based on the region and competition-level in the neighborhood in which the business is located. Just as in 33 Virginia Place, the impact of local competition on sales must be taken into account in any audit scenario.