Statistical Sampling and the New Tangible Property Revenue Procedures
Mary Batcher and Ryan Petska
Federal - 5/15/2012

The Internal Revenue Service’s recent tangible property regulations pose significant challenges for taxpayers.[1] The mechanics of complying with the new rules are addressed in two IRS revenue procedures, Revenue Procedures 2012-19 and 2012-20.[2] This article discusses not the tax aspects of these revenue procedures, but rather the statistical sampling requirements and the implications for taxpayers.

The revenue procedures include several references to the appropriateness of statistical sampling to determine deductions and other tax treatments under the tangible property regulations. Common situations where statistical sampling is expressly recognized as an appropriate methodology include determining repair and maintenance deductions, materials and supplies deductions, and deductions for rotable and spare parts. The statistical sampling approach must be consistent with the guidance provided in Revenue Procedure 2012-42.[3]

Revenue Procedure 2011-42 describes simple random and stratified sampling and lists four estimation procedures as appropriate, albeit with limitations on the use of two of them. In simple random sampling, items are randomly selected until the desired sample size is reached using random numbers to determine whether an item is selected or not. In stratified sampling, the population is divided into buckets or strata and a simple random sample is selected in each stratum. In both types of sampling, the probabilities of selection must be known and nonzero for every item in the population. Revenue Procedure 2011-42 does not authorize the use of non-statistical or judgment sampling, which is characterized by subjective selection of items into the sample.

The statistical sampling language in Revenue Procedures 2012- 19 and 2012-20 is, as follows:

By following the sampling procedures provided in Rev. Proc. 2011-42, 2011-37 I.R.B. 318, a taxpayer changing its method of accounting under section X.XX of the APPENDIX may use statistical sampling in determining the § XXXX. Sampling methodologies not described in Rev. Proc. 2011-42 are not permitted.

The latter quoted sentence has been interpreted by some as allowing only the specific sampling and estimation methods explicitly described in Revenue Procedure 2011-42. Those approaches, however, will not work in all situations and there are other equally, valid types of sampling and estimation that are not described in the sampling revenue procedure. Revenue Procedure 2011-42 includes an acknowledgment that the procedures described are not all encompassing:

It is recognized that existing industry practices and specific taxpayers may be using techniques that are not covered by this revenue procedure. If a taxpayer employed a probability sample or method not covered by this revenue procedure, then the estimate may be referred to a Statistical Sampling Coordinator for resolution or issue development.

Because the tangible property revenue procedures are silent on the acceptability of statistical sampling in many of the situations expressly addressed in the guidance, IRS representatives have been asked to confirm that the IRS did not intend to limit the use of valid statistical sampling in those cases. The IRS has publicly done so, confirming that the propriety of such sampling will depend on the facts and circumstances.[4]

Hence, the door is open to use statistical sampling in tangible property tax determinations. Statistical sampling can help increase efficiency because a single sample can be used for multiple purposes as long as the populations are the same. For example, a sample of buildings can be used to determine repair and maintenance expense and dispositions.

Many companies have used samples to identify qualified repairs deductions or to perform a cost-segregation analysis in the last few years. If those were statistical samples, they can be reanalyzed under the new tangible property rules. The sample can be expanded to incorporate assets placed in service after completion of the original project. Depending on the specific circumstances, the additional sample selections can be relatively small in number.

The IRS has signaled a clear acceptance of statistical sampling in the tangible property revenue procedures. Statistical sampling is a powerful tool that can greatly reduce the cost and burden of compliance with the new regulations and of capitalizing on the potential benefits available. Companies should feel comfortable taking advantage of the savings in time and resources that statistical sampling brings.

Mary Batcher is Executive Director, Quantitative Analysis & Statistics, at Ernst & Young LLP where she leads the statistical sampling practice in Tax. She has successfully directed many sampling projects for tax filing purposes and has worked with the Internal Revenue Service to expand its acceptance of statistical sampling. Ms. Batcher formerly was the chief of a statistical support group at the IRS. She holds a Ph.D. degree in statistics from the University of Maryland, and is an accredited member of the American Statistical Association. An article co-written by Ms. Batcher on Revenue Procedure 201-42 appeared in the September-October issue of The Tax Executive. She may be contacted at

Ryan Petska is a Senior Manager at Ernst & Young. He has more than 10 years of statistical consulting experience in sampling and estimation. While at Ernst & Young, Mr. Petska has worked extensively in tax sampling; prior to that, he served as a statistics instructor at Florida State University. Mr. Petska holds a Master’s degree in statistics from Florida State University, is an accredited member of the American Statistical Association. An article co-written by Mr. Petska on Revenue Procedure 201-42 appeared in the September-October issue of The Tax Executive. He may be contacted at



1. The regulations are the subject of an article in the March-April 2012 issue of The Tax Executive. Dwight N. Mersereau & Corey M. Wise, The New Repair Regulations, 64 Tax Exec. 133 (March-April 2012).
2. See Rev. Proc. 2012-19, 2012-14 I.R.B. (March 7, 2012), and Rev. Proc. 2012-20, 2012-14 I.R.B. (March 7, 2012).
3. See generally Mary Batcher, Ed Cohen, Ryan Petska & Wendy Rotz, New IRS Guidance Boosts Statistical Sampling, 63 Tax Exec. 319 (September- October 2011).
4. For example, Scott K. Dinwiddie, Special Counsel to IRS Associate Chief Counsel (Income Tax & Accounting), made the comments during the ABA Section of Taxation’s May 11, 2012 meeting in Washington, D.C.