RULES AND REGULATIONS
Title 61--REVENUE
DEPARTMENT OF REVENUE
[61 PA. CODE CHS. 6, 8a AND 35]
Enforcement; Tax Examinations and Assessments
[28 Pa.B. 1522] The Department of Revenue (Department), under the authority in section 2910-A of the act of June 30, 1995 (P. L. 139, No. 21) (72 P. S. § 9910-A) (Act 21) and section 270 of the Tax Reform Code of 1971 (TRC) (72 P. S. § 7270), by this order adopts amendments to delete § 6.22 (relating to further examination of books and records); to add Chapter 8a (relating to enforcement); and to amend § 35.1 (relating to tax examinations and assessments).
Purpose of Amendments
The purpose of these amendments is to advise taxpayers of the Department's interpretation of section 2915-A of the TRC (72 P. S. § 9915-A).
Explanation of Regulatory Requirements
Upon final publication of Chapter 8a, the language set forth in § 6.22 is no longer necessary; therefore, the section is being deleted in its entirety.
Section 8a.1 (relating to definintions) defines the following terms for use in this chapter: ''audit period,'' ''block sample,'' ''clustered sample,'' ''deviation from the mean,'' ''outlier,'' ''population,'' ''range,'' ''standard deviation,'' ''standard error,'' ''statistical estimation,'' ''statistical sample,'' ''stratum,'' ''taxpayer,'' ''test audit,'' ''test period'' and ''transaction.''
In accordance with section 2915-A(a) of the TRC, § 8a.2 (relating to examination) provides that the Department may examine all books, papers and records of a taxpayer or another person having possession of or dominion over records to: (1) Verify the accuracy and completeness of a tax return or tax report filed by the taxpayer and ascertain or assess tax or other liability owed to the Commonwealth; (2) Ascertain or assess tax or other liability owed to the Commonwealth if no tax return or tax report has been filed by the taxpayer.
Under section 2915-A(b) of the TRC, § 8a.3 (relating to audit types) provides that examination may be made by desk audit, field audit or another form of audit. Under § 8a.4 (relating to determination of liability), the Department may determine a tax liability owed by a taxpayer based upon the facts contained in a tax return, tax report or other information that may come into the Department's possession.
Section 8a.5 (relating to determination of audit method) provides that when the taxpayer does not have complete records or when the review of each transaction would be unduly burdensome on the Department to conduct an audit in a timely and efficient manner, the Department will determine whether to examine all of the records of a taxpayer for an entire audit period, employ a test audit method or utilize a combination of audit methods. This section lists factors that the Department will consider in determining the audit method.
When a test audit is employed, § 8a.6 (relating to selection of sample) describes the basis for selection of the sample. The Department may utilize stratification levels in performing statistical sampling. When a block sample method is chosen, the Department will select blocks whose average is approximately equal to the estimated average of key characteristics for the audit period. Examples of key characteristics include sales, taxable to gross sales ratio, purchases or number of transactions.
Section 8a.6(1) explains that in determining whether to exclude the values of certain transactions from the sample, the Department will identify the transactions in the sample that are outliers. Outliers are sample values that are so different from the other sample values that it seems unlikely they are representative of the population being audited and, further, whose magnitude is such that including them in the projection could distort the audit findings. Paragraph (2) explains the process for identifying outliers. Paragraph (3) explains the steps that will be taken with respect to confirmed outliers. Paragraphs (4) and (5) detail the factors the Department will consider when determining whether to employ the test audit method in an audit of any Motor Carrier Road Tax and State and Local Sales and Use Tax or Hotel Occupancy Tax or Public Transportation Assistance Tax.
Section 8a.7 (reltaing to statistical estimation and software) provides that the audit results shall be computed by projecting the audit findings identified in the sample, as adjusted for outliers as provided in § 8a.6(3), to the population, regardless of whether the sample is a statistical sample or a block sample. Paragraph (1) states that when the Department employs the block sampling method, the standard error cannot be estimated. Paragraph (2) provides that when the Department employs the statistical estimation method, a standard error of the estimate shall be computed from the sample observations adjusted for outliers as provided in § 8a.6(3) to indicate the reliability of the estimated average, total or ratio. The Department may use software that has been designed in accordance with accepted statistical practices. The formulas utilized by the software will be available for examination by the taxpayer.
Section 8a.7(3) provides that except as otherwise mutually agreed to by the Department and the taxpayer, the number of observations in the sample will be chosen so that the projected sample will, on the average, yield an estimated precision within 25% of the mid-point of a 90% two-sided confidence interval. The sample size will be determined by using the sample size selection table set forth in paragraph (4). Additionally, the Department will increase the sample size upon the request of the taxpayer. The process of increasing the sample size will be repeated until mutual agreement is reached between the taxpayer and the Department on an acceptable number of observations.
Section 8a.8 (relating to test audit plan) provides that prior to conducting a test audit, the Department will set forth in writing a test audit plan and provide the taxpayer with an opportunity to review and comment on the plan. This section further provides areas that the plan will address including the statistical estimationprocedures and the taxpayer's right to request an increase in sample size.
Section 8a.9 (relating to audit findings) provides that at the conclusion of an audit, the Department will provide the audit findings and a copy of the work papers to the taxpayer, discuss the findings with the taxpayer, provide the taxpayer the opportunity to comment in writing and explain the procedure for the processing, assessing and appealing the audit findings. In accordance with section 2915-A(C) of the TRC, § 8a.10 (relating to taxpayer appeal) provides that a taxpayer may appeal the accuracy of a test audit. In accordance with section 2917-A of the TRC (72 P. S. § 9917-A), § 8a.11 (relating to applicability) provides that Chapter 8a applies to all taxes administered by the Department.
Section 35.1 (relating to tax examinations and assessments) is being amended by deleting the current text of subsection (a)(2) because similar language is in Chapter 8a.
Affected Parties
Taxpayers subject to audit by the Department may be affected by these amendments.
Comment and Response Summary
Notice of proposed rulemaking was published at 25 Pa. B. 4005 (September 23, 1995). The amendments are being adopted with changes to the proposed rulemaking.
The Department received comments from the public, the House Finance Committee and the Independent Regulatory Review Commission (IRRC). No objections or comments were raised by the Senate Finance Committee.
Though each comment received raised some unique concerns, many of the comments were similar in nature. Generally, the comments suggested that the proposed amendments did not provide sufficient guidance and detail regarding the use of statistical sampling and test audits. The Department agrees and has incorporated changes suggested by the comments received.
On October 22, 1997, the Department submitted the final-form regulations to IRRC and the Legislative standing committees. At the same time, the Department sent copies of the final-form regulations to the parties who had commented on the proposed rulemaking during the public comment period. Under section 5.1(d) of the Regulatory Review Act, the final-form regulations were deemed approved by the Legislative standing committees on November 12, 1997. On November 20, 1997, IRRC disapproved the final-form regulations.
On December 1, 1997, the Department notified the Governor, IRRC and the Legislative standing committees of its intent to proceed with adoption of the final-form regulations under section 7(a)(2) of the Regulatory Review Act (71 P. S. § 745.7(a)(2)). Under this section and section 7(c) of the Regulatory Review Act, the Department submitted a report to the Legislative standing committees and IRRC with revised final-form regulations. The revised final-form regulations reflect substantial input provided by Legislative committee staff, IRRC staff, public commentators and a private expert statistician contracted by the Department. Numerous telephone calls and drafting meetings were conducted with the identified parties for the purpose of resolving IRRC's concerns.
In response to IRRC's suggestion that a definition section be added, § 8a.1 has been amended to set forth definitions of the following terms: ''audit period,'' ''block sample,'' ''clustered sample,'' ''deviation from the mean,'' ''outlier,'' ''population,'' ''range,'' ''standard deviation,'' ''standard error,'' ''statistical estimation,'' ''statistical sample,'' ''stratum,'' ''taxpayer,'' ''test audit,'' ''test period'' and ''transaction.'' IRRC had also suggested adding the definition of ''Department;'' however, the term ''Department'' is defined in § 1.1 (relating to definitions).
IRRC also suggested that the Department define ''unduly burdensome'' when used in the phrase ''unduly burdensome on the Department to conduct an audit in a timely and efficient manner'' proposed in § 8a.1(d), now § 8a.5. It is the Department's position that the term cannot be defined because each case presents unique fact situations that must be considered individually. No one standard can be applied to all taxpayers. The Department has provided in § 8a.5 a listing of considerations that will be considered in determining the type of audit method to be employed.
In response to IRRC's request that the Department clarify proposed § 8a.1(a), § 8a.2 contains two paragraphs.
A public comment expressed concern that because section 2915-A was included in Article XXIX-A, Tax Amnesty Program, the definitions in section 2901-A of the TRC also apply to section 2915-A. Following this theory, the provisions of section 2915-A of the TRC and Chapter 8a would only apply to a taxpayer participating in the Tax Amnesty Program with regard to certain specified eligible taxes delinquent as of December 31, 1993. However, section 2917-A of the TRC specifically states that section 2915-A of Article XXIX-A shall apply to all taxes collected by the Department. Because the Legislature did not use the defined term ''eligible tax'' in this section, section 2915-A clearly applies to all taxes collected by the Department, not just ''eligible taxes'' under the Tax Amnesty Program.
The language proposed in § 8a.1(e) has been deleted and replaced by § 8a.5 with language that explains various factors the Department will consider to determine whether to examine all of the records of a taxpayer for an entire audit period, employ a test audit method or utilize a combination of audit methods.
In the final-form regulations disapproved by IRRC, the Department redrafted § 8a.5 (formerly proposed § 8a.1(e)) related to the list of factors the Department may consider in determining whether to conduct a complete audit, a test audit or a combination of audit methods. Section 8a.6(3) and (4) (formerly proposed subsection (f)(3) and (4)) was also amended to advise taxpayers of additional factors that the Department may consider in determining whether to conduct a test audit in Motor Carriers Road Tax, Sales Tax, Use Tax and Hotel Occupancy Tax. In the proposed rulemaking, these subsections had provided that the Department will consider these lists of factors in selecting an appropriate audit method.
IRRC has indicated that the distinction between the terms ''may'' and ''will'' in these sections is significant. The Department's sole purpose for creating these lists was to identify for taxpayers the types of factors that would be considered by the Department in its selection of an audit method. In conformity with IRRC's concerns, the Department has revised former subsections (e), (f)(3) and (f)(4) to provide that the Department will consider the identified factors.
In its disapproval order, IRRC also indicated that the final-form regulations do not provide assurance that theliability determined by the Department using statistical sampling or test audits will be accurate within any degree of precision. IRRC also cited a letter from the majority and minority Chairpersons of the House Finance Committee dated November 18, 1997, which provides that the final-form regulations do not sufficiently address the risk of over-assessment. This comment suggested the use of the lower limit of either a 90% two-sided confidence interval or a 95% one-sided confidence interval, both of which are identical. The rationale for using the lower limit of these confidence intervals is that it would reduce the risk of over-assessment to no more than 5%.
After extensive review of this issue, it is the Department's opinion that it is inappropriate to use the lower limit of a confidence interval to determine an audit finding. The lower limit of a confidence interval is a very unlikely value for the true value. While it is true that the use of the recommended lower limit would reduce the risk of over-assessment to no more than 5%, it would also increase the risk of under-assessment to no less than 95%.
The Department believes that the use of the lower limit of a confidence interval is not in the best interests of the accurate enforcement of the Commonwealth's tax laws. The Department also believes that the use of the lower limit is unfair to taxpayers that have accurately reported their liabilities. The midpoint of the confidence interval is the most accurate estimate of the true value of the audit finding. Therefore, the Department's regulations utilize the midpoint of a 90% confidence interval to determine the audit finding.
Related to this concern is the issue of the level of precision that should be utilized in statistical estimation. Precision as used in the amendments is the range within which the average value will lie, with the degree of certainty specified in the confidence interval. Although IRRC and the Legislative standing committees have made no formal recommendation of an acceptable precision, discussions with staff and public commentators have suggested precisions ranging from 5 to 20%. Surveys of the practices of other states indicate that the precisions routinely used by State tax agencies in test audits range from 5 to 50%. In addition, many states do not calculate the precision of their test audits.
The Department has given a great deal of consideration to the establishment of a minimum precision level. Although a high precision (for example, 5%) may be an ideal goal, the Department believes it is not appropriate to mandate an extremely high precision level for the selection of an initial sample for the following reasons:
First, the precision of a sample as measured by the confidence interval cannot be estimated without first knowing the standard deviation or coefficient of variation of the sample. The data to be projected in tax audits is highly variable by nature and is constantly changing due to frequent statutory amendments and changes in business practices. This limits the Department's ability to estimate the coefficient of variation of a sample to be selected based upon historical data.
Therefore, it is the Department's position that the best method for estimating the precision of a sample in a tax audit is to select an initial sample and calculate its coefficient of variation and precision. This process provides a basis for making a more accurate estimation of the precision to be achieved by any additional sample selected. If the taxpayer requests an increase in the number of observations being reviewed, the sample size can be increased. However, if the initial sample selected is satisfactory to both the taxpayer and the Department, there is no necessity to mandate that additional samples be selected.
Second, the use of stratification by the Department in conducting audits limits the potential range of the taxpayer's liability determined in a test audit. In a stratified audit, the transactions being audited are subdivided into several homogenous groups with respect to the characteristics being audited. For example, the transactions may be subdivided by dollar amount. In conducting a stratified audit, the Department may elect to do a complete audit on the subdivided groups containing the transactions with the largest dollar values. However, on the small dollar value groups, the Department may elect to use statistical estimation.
The Department has documented examples of its stratified audits when a review of only 5% of the total transactions in the sample resulted in the actual examination by the Department of over 50% of the total gross receipts that were the subject of the audit. Because complete audits were done on the transactions with the highest dollar values, and greatest impact on the taxpayer's liability, a low precision in the small dollar transaction strata may not significantly affect the taxpayer's ultimate tax liability.
Finally, it is the experience of the Department that many taxpayers do not want the Department to examine the number of samples required to obtain extremely high precision levels. This is due to the fact that the taxpayer must search for and identify the record for every sample transaction to be included within the projection. These records may be located in various facilities across the country and difficult to locate.
In addition, the taxpayer may be required by the Department to answer questions and provide additional verifications to support the records selected in the sample. This often requires the taxpayer to search for and identify additional records and identify the employes that were involved in the questioned transactions. These records and employes may also be located throughout the country. If the relevant employes have left the company, the taxpayer's reconstruction and verification of the record is made even more difficult.
In the case of stratified audits, the taxpayer's cost of pulling records related to and justifying large numbers of relatively small dollar transactions routinely outweighs any justification for an extremely high precision level. Therefore, the burden imposed on the taxpayer in selecting large sample sizes is often greater than the burden imposed on the Department to review the samples selected.
Accordingly, in response to the concerns on reasonableness and clarity of the procedures used for statistical sampling and test audits, the Department has revised the procedure in § 8a.7(3) for determining the number of observations to be selected in the sample. The revised procedure provides that the taxpayer and the Department may mutually agree on the number of observations to be chosen prior to conducting any sampling. In the absence of an agreement, the initial sample selected by the Department will be chosen so that the projected sample will on average yield an estimated precision within 25% of the midpoint of a 90% two-sided confidence interval.The midpoint of a confidence interval is the best estimate of the population characteristic. Section 8a.7(4) provides a sample size selection table to be used in determining the number of observations to be selected in the initial sample.
After the selection and review of the sample, the standard error and estimated precision of the sample will be calculated and reviewed with the taxpayer. Upon the request of the taxpayer the size of the sample will be increased. The process of increasing the sample size will be repeated until mutual agreement is reached between the taxpayer and the Department on an acceptable number of observations. The Department believes that this process mitigates the concerns raised regarding a minimum precision level.
The Department has also significantly revised § 8a.6 in the revised final-form regulations relating to the procedures for the identification and treatment of outliers. Outliers are extreme values that are contained within the sample that are atypical of the population being audited.
The revised procedures require the Department to determine the transaction difference for each transaction in the sample. The transaction difference is the difference between the audited value of the transaction and its value reported to the Department. If a transaction difference is greater than 2% of the total audited amount of the total sample, the transaction is a suspected outlier. A confirmation test is then completed for each suspected outlier using a mathematical formula contained within § 8a.6(2). The test for an outlier is based upon the difference between the value of the outlier (either positive or negative) and the average of all other sampled values, divided by a measure of the dispersion of the other values (1/4 of the range).
Revised § 8a.6(3) provides that the Department will notify the taxpayer of all confirmed outliers and request evidence that would justify a smaller difference between the audited value and the reported value. If sufficient evidence is not provided, the outlier will be eliminated from the sample and audited independently. The audit finding on the outlier will be added to the result of the projection for the remaining sample to determine the total audit finding.
Excluding outliers from the sample projected and auditing them separately should on average yield a tax deficiency that is smaller than if a complete audit of all transactions were used to determine the tax liability owed to the Commonwealth. This results from the fact that there may be other extreme values in the population that are not included in the projections since they were not represented in the sample. This procedure facilitates the recommendation of IRRC and the Legislative standing committees that the audit procedures should minimize the risk of over-assessment.
In the final-form regulations disapproved by IRRC, § 8a.6 provided that:
When a test audit method is chosen to reduce burden, or because certain records are unavailable, or for any other reason, the concurrence of the taxpayer in the test audit plan will be sought. In the absence of concurrence of the block sampling method, the Department will select blocks.... (Emphasis added.)IRRC identified two clarity concerns with this provision. First, IRRC stated that the phrase ''or for any other reason'' should be removed from the final-form regulation because it lacks clarity and does not track the statutory language. The Department agrees and has removed the questioned language from the final-form regulations.
Second, IRRC stated that the phrase ''In the absence of concurrence of the block sampling method'' is confusing and lacks clarity. In response to comments made by IRRC at the public meeting on the final-form regulations and comments directed to IRRC by a public commentator requesting the removal of the references to the concurrence of the taxpayer, the Department has revised this section by deleting the references to the concurrence of the taxpayer.
Section 8a.7(2) is amended by deleting the term ''generally'' from the phrase ''generally accepted statistical practices.'' It was brought to the attention of the Department by public commentators that the phrase has not been defined by any organization of expert statisticians or auditors. Therefore, it was concluded that the deletion of the term ''generally'' from the phrase did not change its meaning and the Department agreed to make the recommended change.
In response to a comment made by the public, IRRC and the House Finance Committee, proposed subsection (h), now § 8a.8 has been amended to provide that prior to conducting a test audit, the Department will set forth in writing a test audit plan and provide the taxpayer with an opportunity to review and comment on the plan. The section sets forth areas that the plan will address including the statistical estimation procedures and the taxpayer's right to request an increase in sample size.
A public comment suggested that if the Department determined a tax liability based on information outside of the tax return or tax report, that it will provide a copy of the information to the taxpayer for purposes of determining the accuracy of the information. The Department has responded to this comment in § 8a.9 by providing that at the conclusion of the audit, the audit findings and a copy of the work papers will be provided to the taxpayer. In addition, the auditor will also discuss the findings with the taxpayer, provide the taxpayer the opportunity to comment in writing and explain the procedures for the processing, assessing and appealing of the audit findings.
In response to a concern raised by IRRC and in accordance with section 2915-A(C) of TRC, a new § 8a.10 (relating to taxpayer appeal) provides that a taxpayer may appeal the accuracy of a test audit by providing clear and convincing evidence that the method used for selecting a statistical sample or block sample test period and determining the tax liability is erroneous, lacks a rational basis or produces a different result when the complete records are considered.
Finally, to avoid any conflict or confusion, the Department is amending § 35.1 by deleting the current text of subsection (a)(2) because similar language is now contained in Chapter 8a. New language has been added to subsection (a)(2) that states that audits will be conducted in accordance with Chapter 8a.
Fiscal Impact
The Department has determined that the amendments will have no significant fiscal impact on the Commonwealth.
Paperwork
The amendments will not generate significant additional paperwork for the public or the Commonwealth.
Effectiveness/Sunset Date
The amendments will become effective upon final publication in the Pennsylvania Bulletin. The amendments are scheduled for review within 5 years of final publication. No sunset date has been assigned.
Contact Person
The contact person for an explanation of the amendments is Anita M. Doucette, Office of Chief Counsel, Department of Revenue, Dept. 281061, Harrisburg, PA 17128-1061.
Regulatory Review
Under section 5(a) of the Regulatory Review Act (71 P. S. § 745.5(a), on September 13, 1995, the Department submitted a copy of the notice of proposed rulemaking, published at 25 Pa.B. 4004, to IRRC and the Chairpersons of the House Committee on Finance and the Senate Committee on Finance for review and comment. In compliance with section 5(b.1) of the Regulatory Review Act, the Department also provided IRRC and the Committees with copies of all comments received, as well as other documentation.
Under section 5(c) of the Regulatory Review Act, IRRC and the Committees were provided with copies of the comments received during the public comment period, as well as other documents when requested. In preparing these final-form regulations, the Department has considered all comments received from IRRC, the Committees and the public.
Under section 5.1(d) of the Regulatory Review Act (71 P. S. § 745.5a(d)), these final-form regulations were deemed approved by the Committees on January 15, 1998. Under section 5.1(e) of the Regulatory Review Act, IRRC met on January 29, 1998, and approved the final-form regulations.
Findings
The Department finds that:
(1) Public notice of intention to adopt the amendments has been given under sections 201 and 202 of the act of July 31, 1968 (P. L. 769, No. 240) (45 P. S. § 1201 and 1202) and the regulations thereunder, 1 Pa. Code §§ 7.1 and 7.2.
(2) The final-form regulations are necessary and appropriate for the administration and enforcement of the authorizing statute.
Order
The Department, acting under the authorizing statute, orders that:
(a) The regulations of the Department, 61 Pa.Code, are amended by deleting § 6.22; amending § 35.1 and adding §§ 8a.1--8a.11 to read as set forth in Annex A.
(b) The Secretary of the Department shall submit this order and Annex A to the Office of General Counsel and the Office of Attorney General for approval as to form and legality as required by law.
(c) The Secretary of the Department shall certify this order and Annex A and deposit them with the Legislative Reference Bureau as required by law.
(d) This order shall take effect upon publication in the Pennsylvania Bulletin.
ROBERT A. JUDGE, Sr.,
Secretary(Editor's Note: For the text of the order of the Independent Regulatory Review Commission relating to this document, see 28 Pa.B. 859 (February 14, 1998).)
Fiscal Note: 15-371. No fiscal impact; (8) recommends adoption.
Annex A
TITLE 61. REVENUE
PART I. DEPARTMENT OF REVENUE
Subpart A. GENERAL PROVISIONS
CHAPTER 6. TAX AMNESTY PROGRAM § 6.22. (Reserved).
CHAPTER 8a. ENFORCEMENT Sec.
8a.1. Definitions. 8a.2. Examination of books and records. 8a.3. Audit types. 8a.4. Determination of liability. 8a.5. Determination of audit method. 8a.6. Selection of sample. 8a.7. Statistical estimation and software. 8a.8. Test audit plan. 8a.9. Audit findings. 8a.10. Taxpayer appeal. 8a.11. Applicability. § 8a.1. Definitions.
The following words and terms, when used in this chapter, have the following meanings, unless the context clearly indicates otherwise.
Audit period--The period of time for which the audit is conducted.
Block sample--One or more groups of transactions selected as a unit from a population. For example, invoices numbered 100 to 200, or transactions for the months of May and October.
Clustered sample--A statistical sample in which blocks of adjacent transactions are selected with known probability. A statistical sample of transactions within the blocks may be selected, creating a two-stage statistical sample.
Deviation from the mean--The numerical difference between a single statistical observation and the mean (average) of all of the statistical observations.
Outlier--A statistical observation that appears to deviate markedly from other members of the sample from which it came.
Population--The total transactions during an audit period from which the sample is selected.
Range--The numerical difference between the largest and smallest statistical observations in the sample.
Standard deviation--The square root of the average squared deviation from the mean.
Standard error--The standard deviation divided by the square root of the number of statistical observations in the sample.
Statistical estimation--A method of estimating the numerical characteristics of a population, such as averages, totals or ratios, from a statistical sample and estimating the precision of the estimated characteristics.
Statistical sample--A selection of transactions in which each of the transactions in the population, or a stratum from it, has a known chance of being selected. The term is also known as a probability sample.
Stratum--A subdivision of the population in which the transactions within the subdivision are expected to be more uniform with respect to the characteristics being examined than the transactions across the subdivisions.
Taxpayer--A person, association, fiduciary, partnership, corporation or other entity required to pay, withhold or collect any tax that is administered by the Department.
Test audit--An audit of sampled transactions selected by either a block sample or a statistical sample method.
Test period--A time period or periods selected for the test audit; for example, the month of May.
Transaction--The term includes an entry, document, invoice or other record regardless of the method of creation or retention.
§ 8a.2. Examination of books and records.
The Department may examine all books, papers and records of a taxpayer or another person having possession of or dominion over these records to:
(1) Verify the accuracy and completeness of a tax return or tax report filed by the taxpayer and ascertain or assess tax or other liability owed to the Commonwealth.
(2) Ascertain or assess tax or other liability owed to the Commonwealth if no tax return or tax report has been filed by the taxpayer.
§ 8a.3. Audit types.
Examination may be made by desk audit, field audit or another form of audit.
§ 8a.4. Determination of liability.
The Department may determine tax liability owed by a taxpayer to the Commonwealth based upon the facts contained in a tax return, a tax report or other information that may come into the Department's possession.
§ 8a.5. Determination of audit method.
When the taxpayer does not have complete records or when the review of each transaction would be unduly burdensome on the Department to conduct an audit in a timely and efficient manner, the Department will determine whether to examine all of the records of a taxpayer for an entire audit period, employ a test audit method or utilize a combination of audit methods. In making this determination, the Department will consider the following factors:
(1) The type of tax under audit.
(2) The nature of the taxpayer's business.
(3) The number of transactions in the population.
(4) The adequacy and availability of the taxpayer's records.
(5) Whether the taxpayer's business is cyclical or seasonal.
(6) Whether significant changes in the taxpayer's business or activities occurred during the audit period, such as discontinuing or adding a line of business.
(7) Other relevant factors.
§ 8a.6. Selection of sample.
When a test audit is employed, the selection of the block sample, statistical sample or clustered sample shall be based on the Department's analysis of the taxpayer's business operations and records, and shall reasonably represent the population from which the sampled transactions were selected. The Department may utilize stratification levels in performing statistical sampling. When a block sample method is chosen, the Department will select blocks whose average is approximately equal to the estimated average of key characteristics for the audit period. Examples of key characteristics include sales, taxable to gross sales ratio, purchases or number of transactions.
(1) In determining whether to exclude the values of certain transactions from the sample, the Department will identify the transactions in the sample that are outliers.
(2) For the purpose of identifying outliers, the Department will determine the transaction difference for each transaction in the sample. The transaction difference shall be the difference between the transaction's audited value and its value reported to the Department. Any transaction difference with an absolute value greater than 2% of the total audited amount of the total sample shall be considered to be a suspected outlier. If the difference is no greater than 2% of the total audited amount of the total sample, no adjustment will be made. If the difference is greater than 2% of the total audited amount of the total sample, the following test will be done: Subtract the average of the transaction differences, omitting the suspected outlier, from the suspected outlier and divide by one-fourth of the range in values of the transaction differences, omitting the suspected outlier. If the absolute value of the ratio is four or greater, the suspected outlier shall be confirmed as an outlier. If there is more than one suspected outlier, this test shall be applied sequentially to all suspected outliers. If the population is stratified this process will be completed for each stratum in which sampling has been done.
(3) The following steps will be taken with respect to all confirmed outliers:
(i) The taxpayer will be notified concerning the outliers and requested to furnish evidence that will be considered by the auditor in determining the audited finding. If, upon examining the further evidence, the auditor agrees that a smaller difference between the reported amount and the audited amount is justified, the auditor will replace the original transaction by the adjusted finding.
(ii) If sufficient evidence is not provided, the outlier will be eliminated from the sample and audited independently. The audit finding on the outlier will be computed separately and the audit finding will be added to or, if negative, subtracted from the result of the projection for the remaining sample.
(iii) The sample values, adjusted for outliers as provided in subparagraphs (i) and (ii), will be used for projection of the total audit finding and its standard error.
(4) When determining whether to employ the test audit method in an audit of a tax under 75 Pa.C.S. §§ 9601--9622, (relating to motor carriers road tax) or a similar tax which may be enacted, the Department will consider the following factors:
(i) The average fleet mileage as reported by the taxpayer.
(ii) Whether the vehicles are company-owned, permanently leased from owner-operators, or a combination of both.
(iii) The types of vehicles that make up the fleet.
(iv) The type of fuel used to power the vehicles.
(v) The geographical area in which the vehicles operate.
(vi) The type of commodities being hauled.
(vii) The total number of vehicles in the taxpayer's fleet.
(viii) The adequacy and availability of the taxpayer's records.
(ix) Whether the taxpayer's business is cyclical or seasonal.
(x) Whether significant changes in the taxpayer's business or activities occurred during the audit period, such as discontinuing or adding a line of business.
(xi) Other relevant factors.
(5) When employing a test audit method in an audit of a State or local Sales and Use Tax or Hotel Occupancy Tax or Public Transportation Assistance Tax (72 P. S. §§ 7201--7282 and 9301; 53 P. S. §§ 12720.501--12720.509; 16 P. S. §§ 6150-B--6157-B) or a similar tax which may be enacted, the Department will consider the following factors:
(i) The average gross sales.
(ii) The ratio of taxable sales to gross sales.
(iii) Whether the taxpayer's business is cyclical or seasonal.
(iv) Whether significant changes in the taxpayer's business or activities occurred during the audit period, such as discontinuing or adding a line of business.
(v) The adequacy and availability of the taxpayer's records.
(vi) Other relevant factors.
§ 8a.7. Statistical estimation and software.
The audit results shall be computed by projecting the audit findings identified in the sample, as adjusted for outliers as provided in § 8a.6(3) (relating to selection of sample) to the population, regardless of whether the sample is a statistical sample or a block sample.
(1) When the Department employs the block sampling method, the standard error cannot be estimated.
(2) When the Department employs the statistical estimation method, a standard error of the estimate shall be computed from the sample observations adjusted for outliers as provided in § 8a.6(3) to indicate the reliability of the estimated average, total or ratio. The Department may use software that has been designed in accordance with accepted statistical practices. The formulas utilized by the software will be available for examination by the taxpayer.
(3) Except as otherwise mutually agreed to by the Department and the taxpayer, the number of observations in the sample will be chosen so that the projected sample will, on the average, yield an estimated precision within 25% of the midpoint of a 90% two-sided confidence interval. In determining the size of the sample, the Department will use the sample size selection table in paragraph (4). The estimated precision of the sample selected may be less than or greater than 25%, depending upon the variability in the sample data. The standard error and estimated precision will be calculated and reviewed with the taxpayer. The sample size will be increased upon the request of the taxpayer. The process of increasing the sample size will be repeated until mutual agreement is reached between the taxpayer and the Department on an acceptable number of observations.
(4) The following sample size selection table identifies estimated sample sizes required to produce estimates with specified precision:
Sample Size Selection Table
Precision Confidence Estimated Coefficient of Variation (CV)
Interval (2 sided) 0.25 0.50 0.65 0.75 1.00 1.50 2.00 3.00 5.00 5% 90% Normal 68 271 650 609 1,083 2,435 4,330 9,742 27,060 10% Deviate * 68 115 153 271 609 1,082 2,435 6,765 15% 1.645 * 31 51 68 121 271 481 1,082 3,007 20% * * 29 39 68 152 271 609 1,691 25% * * * 25 44 97 173 390 1,082 30% * * * * 31 68 120 271 752 35% * * * * 23 50 88 199 552 40% * * * * * 38 68 152 423 * Fewer than 20 sample observations are required.
§ 8a.8. Test audit plan.
Prior to conducting a test audit, the Department will set forth in writing a test audit plan and provide the taxpayer with an opportunity to review and comment on the plan. The plan will describe the time period subject to audit, the records subject to review, methods for selecting records, statistical estimation procedures including the taxpayer's right to request an increase in sample size and the manner in which any tax liability will be calculated based upon the records reviewed.
§ 8a.9. Audit findings.
At the conclusion of the audit, the audit findings and a copy of the work papers will be provided to the taxpayer. The auditor will:
(1) Discuss the findings with the taxpayer.
(2) Provide the taxpayer the opportunity to comment in writing.
(3) Explain the procedures for the processing, assessing and appealing of the audit findings.
§ 8a.10. Taxpayer appeal.
The taxpayer may appeal the accuracy of a test audit by providing clear and convincing evidence that the method used for selecting a statistical sample or block sample test period and determining the tax liability is erroneous, lacks a rational basis or produces a different result when the complete records are considered.
§ 8a.11. Applicability.
This chapter applies to all taxes administered by the Department.
ARTICLE II. SALES AND USE TAX
CHAPTER 35. TAX EXAMINATIONS AND ASSESSMENTS § 35.1. Tax examinations and assessments.
(a) Examinations. Tax examinations shall conform with the following:
* * * * * (2) Audits. Audits shall be conducted in accordance with Chapter 8a (relating to enforcement).
[Pa.B. Doc. No. 98-481. Filed for public inspection March 27, 1998, 9:00 a.m.]
No part of the information on this site may be reproduced for profit or sold for profit.This material has been drawn directly from the official Pennsylvania Bulletin full text database. Due to the limitations of HTML or differences in display capabilities of different browsers, this version may differ slightly from the official printed version.
