What Does the Future Hold For At-Cost Services Charges?

Introduction

The Internal Revenue Service's new regulations under section 482 relating to intercompany services were intended to simplify the markup question for intercompany charges for routine services going forward. Have they? One of the biggest concerns many taxpayers will have with the

new regulations is the administrative burden of charging their services at cost from year-to-year. The research summarized in this article shows that the Services Cost Method (SCM) set forth in the new regulations will not fully meet the expectations of taxpayers since one of its primary tests can fail in some years and not in others, simply because of changes in the business cycle. This creates heightened administrative burdens and audit risks that were likely not envisioned upon creation of the SCM.

As background, the final and temporary regulations on the treatment of services under section 482 provide taxpayers with the ability to charge intercompany services at cost to foreign affiliates under the Services Cost Method (SCM). In order to qualify for a charge out at cost under the SCM, the taxpayer must reasonably conclude in its "Business Judgment"1 that "the performed services do not contribute significantly to key competitive advantages, core capabilities, or fundamental risks of success or failure in the business of the renderer, recipient, or both."2 In other words, the performed services must be deemed "routine" by the taxpayer.

Additionally, the taxpayer must: (i) match up the services against a list of specified services that may be charged at cost3; or (ii) have a median arm's-length markup on total services costs of less than or equal to seven percent (the seven-percent threshold).4 The second criterion was created because the 1RS recognized that the list of specified covered services might not encompass all categories of routine services performed by taxpayers.5

The article focuses on the seven-percent threshold and whether it provides taxpayers with a stable decision point to charge services at cost or with a markup throughout time. Specifically, it reflects an analysis of statistics related to U.S. services companies for nearly 206 years and evaluates whether there are any trends in the median markups of these services companies over time that may affect the seven-percent threshold under the new services regulations for taxpayers.

Analysis

The process again with an identification of the Standard Industrial Classification (SIC) codes that capture services companies similar to the typical routine management service functions performed by a multinational taxpayer in the United States. As part of this process, the SIC codes were aligned with the activities that frequently overlap with the categories and activities that encompass the Specified Covered Services list under the IRS's services regulations.7 By combining a departmental approach from the taxpayer's perspective and the IRS's services regulations, a process was developed that identified the SIC codes that align with typical routine management service departments (e.g., tax, accounting, etc.). These are departments that many taxpayers would deem to be low margin or routine and would need to test if the median markup is less than or equal to seven percent to administer an at-cost charge under the SCM.

The SIC code search resulted in a variety of codes that identified virtually all U.S. services companies that may provide low-margin service functions.8 The selected SIC codes lead to the identification of the first group, called "AU Service Codes," that captures virtually all the services companies that perform activities similar to a multinational taxpayer's back office including: accounting, tax, legal, finance, human resources, treasury, G?, and marketing. This group also overlaps with the majority of the categories listed under the Specified Covered Services within the IRS's services regulations. An analysis of All Services Codes permits an assessment of the entire services' industry returns and median markups over the long-run (i.e., a 19-year period).

Next, the codes were refined to formulate two additional groups:

* All Corporate Service Codes - SIC codes that identify a group of companies that perform activities within each of the following departments: accounting, tax, legal, finance, and human resources; and

* All Marketing Service Codes - SIC codes that identify a group of companies that perform marketing services.

By separately analyzing the All Corporate Service Codes and All Marketing Service Codes groups, the cost markup trends can be examined as they would apply to the activities of certain departments over time.

For the fourth group, All Corporate Service Codes was further segmented into a select group of ten companies that represent the Specified Corporate Service Companies group. The process of identifying a select group of companies, similar to the Specified Corporate Service Companies, is common for a transfer pricing analysis. In fact, this level of granularity would be expected under the new services regulations to evaluate an at-cost charge under the SCM. The Specified Corporate Service Companies group covers the routine functions performed by typical accounting, tax, legal, finance, and human resources departments.

In order to develop the Specified Corporate Service Companies, a review was conducted of business descriptions, SEC filings (e.g., 10-Ks, 8-Ks, and 10Qs), and websites (complemented by telephone interviews to each comparable company) to ensure they performed routine functions that aligned with common accounting, tax, legal, finance, and human resource departments. The result was a set of ten companies that align with these departments that a multination- al taxpayer would categorize as a low margin, non-routine, function and desire to charge at cost.1*

Once the four groups were assembled, a rate-of-return analysis was performed for the groups that covered all SIC codes, from 1987 to 2005. The rate of return was measured as the return on total operating costs'" (ROTOC) for each available company from 1987 to 2005. The rate-of-return analysis was used to testing a broad sample of economic data for services companies rather than a small selection of comparables.11

Next, the data on a single year and multi-year analysis were analyzed and the median markup over a 19-year period was plotted. The time period of 1987 to 2005 was used for the companies that constitute All Service Codes, All Corporate Service Codes, and All Marketing Service Codes. Regrettably, for the Specified Corporate Service Companies, the time period was constricted to 1995 to 2005 because reliable data for the comparable companies were not available prior to 1995 and were not included in this analysis.

The purpose of the single-year analysis is to illustrate the direct variations in the median markup for each group over a 19-year period. In addition, a single-year analysis is helpful in illustrating how various market factors may influence the returns of an entire service group (and industry) for any given year. Also, the seven-percent threshold over this entire period is highlighted to facilitate a comparison of each group's results. This analysis is illustrated in Figure 1 (on page 256), which shows that the median markup for three of the groups of services exceeded the seven percent threshold from 1987 to 2005. All Service Codes exemplified the lowest amount of variability, which is to be expected given the broad grouping of companies included in this group.

For the multi-year approach, the data were analyzed on a simple three-year average median markup. The new services regulations do not define the number of years that should be used to determine if a services charge can be made at cost under the SCM. Experience suggests that the most common approach used by most taxpayers is« three-year average, which is largely due to common practices by transfer pricing consultants and the example in Treas. Reg. §1.482-1(0 that deals with multi-year averages, aggregation, and other general principles that might be needed to evaluate an arm's-length result in a particular case.

A three-year analysis is very helpful because it illustrates the median markup variations that have occurred over short-term economic cycles and how this would affect the median markups if it were performed under the SCM for that year. As with the single year analysis, the seven-percent threshold was highlighted over this entire period to evaluate each group's comparison. This analysis is illustrated in Figure 2 (on page 256), which reveals that the median markup for two of the groups of services exceeded the seven percent threshold from 1987 to 2005. In this instance, AU Service Codes and AU Corporate Services Codes exemplified the lowest amount of variability.

The returns to the services companies illustrated in Figures 1 and 2 should be mean-reverting time series. Said differently, the returns to the services companies should vary about a long-run central point of tendency as the economy undergoes various cycles. The long-run mean, or central point of tendency, was calculated as the average median return, but because reliable data for the Specified Corporate Services Companies were only available from 1995 to 2005, the long-run mean was calculated only for this period. Table 1 provides a summary of the long-run means for each group of services companies.

IMAGE TABLE1

Table 1 : Long-run Means for Services Groups

Table 1 demonstrates that the long-run means for the first three groups of services are between 2.5 percent and 3.2 percent from 1987 to 2005, significantly below the seven-percent threshold under the SCM. In contrast, the long-run mean for the Specified Corporate Services Companies is 6.9 percent from 1995 to 2005, approximately the same as seven-percent threshold under the SCM. What does this tell us? The long-run means are only one part of the picture and must be analyzed in conjunction with the volatility of services companies.

Observations

The fluctuations in each group over the time period were surprising, leading to the following observations and conclusions regarding the future of seven-percent threshold under the SCM for taxpayers:

* Volatility of Services Companies: The seven-percent threshold under the SCM may not be stable for taxpayers given the volatility of the returns illustrated in Figures 1 and 2. This is especially true for the AU Marketing Codes12 and Selected Corporate Services Companies, which both exceeded the seven-percent threshold during the tested periods. This seems somewhat counter-intuitive given that one objective of the SCM was to simplify taxpayers' decisions to markup intercompany charges for routine services.

This indicates that taxpayers using a select group of comparables under the SCM may have to consider a markup in some years when the median return for this group exceeds the seven-percent threshold because of volatility in the markups of services companies.

* Use of Multi-year Data: The use of single and multiple-year data can have an effect on whether the comparables fall under the seven-percent threshold. Because the new services regulations do not define the number of years that should be used to determine if a services charge can be made at cost under the SCM, a single-year and multi-year approach will provide varying results. There may even be situations where a single year analysis is above seven-percent and the multi-year approach is below.

Taxpayers should create a transfer pricing policy for administering at-cost charges under the SCM that will provide them flexibility and minimizes the impact that a one-year anomaly may have on the outcome of their analysis. By utilizing three or more years when performing the SCM analysis, a taxpayer could help mitigate the risk that their comparables would surpass the seven percent threshold.

* Selection of Comparables: Because of the volatility of services companies, taxpayers should carefully identify comparable services companies that align with the functions performed by their corporate departments. If the comparables are not carefully selected, the ability to make at-cost charges under the SCM may vary depending on the current (and past) economic cycles and the resulting returns for the specific services companies selected for calculation of an arm's length markup. For example, the All Marketing Codes, All Corporate Codes, and Selected Corporate Service Companies surpassed the seven-percent threshold in the single-year analysis. Only the All Marketing Codes and Selected Corporate Service Companies surpassed the seven-percent threshold in the multi-year analysis. Again, this stresses the importance of evaluating each company to ensure it is comparable to the routine functions being performed to qualify for an at-cost charge under the SCM.

* Contemporaneous Analysis: When a return to a routine service is close to the seven-percent threshold, taxpayers may need to consider refreshing the comparables annually. Because the analysis reveals significant variation in the median markups from 1987 to 2005, an update analysis should be performed annually to evaluate these market changes for the company's transfer pricing policy. This will help mitigate potential 1RS audit and adjustment exposures.

* Seven Percent Threshold: Without knowing the process used by the 1RS in determining the seven-percent threshold, it is unclear how it was established and whether will it stand the test of time. It is ironic that the period in which the seven-percent threshold was likely created coincided with the lowest returns (2001 - 2003) observed amongst the four groups analyzed because of the dot-com bubble, September 11th, and several major corporate accounting scandals. Perhaps the threshold should be modified to incorporate the peaks and valleys of the services companies. Perhaps the threshold should vary and be recalculated on an annual basis to properly reflect the returns experienced by the services industry. For now, taxpayers should consider how their comparables will endure with the current seven-percent threshold under the SCM.

Conclusion

The future of the seven-percent threshold under the SCM is uncertain. While the intention of the new services regulations was to simplify intercompany charges for routine services, the rate-ofrerurn analysis summarized in this article suggests that this may not be the case. Furthermore, when utilizing the seven-percent threshold under the SCM, taxpayers may be in a position where the median markup moves above seven-percent. In this situation, the taxpayer's auditor or the 1RS may challenge the characterization of the routine service; perhaps leading to future markups and challenges to the historical at-cost charges.

Needless to say, careful consideration must go into the comparables analysis, or the taxpayer may end up charging at cost in one year and then with a markup in next. One thing that is for certain is that taxpayers will likely have to continue to perform comparables analyses unless their intercompany services match the list of specified services that may be charged at cost, which is an administrative burden perhaps not intended by the new services regulations.

IMAGE GRAPH2

Figure 1:

Median Yearly Return on Total Operating Costs (ROTOC)

Figure 2:

Median ThreeYear Average Return on Total Operating Costs (ROTOC)

REFERENCE

1. Under Notice 2007-03 (January 16, 2007), services must be analyzed using the business judgment rule, but a taxpayer may use the existing or temporary services regulations to analyze each intercompany service transaction.

2. Temp. Reg. § 1.482-9T(b)(2).

3. See Notice 2007-5 and Revenue Procedure 2007 1 3.

4. Temp Reg. § 1.482-9T(b)(4)(ii).

5. 15 BNA Transfer Pricing Report 338 (September 13, 2006) ("Branch 6 Chief Clarifies Application of Cost Method, Other Aspects of New Temporary Services Regulations").

6. The Standard & Poor's Compustat database used for this article has data that goes back to 1987; thus, 19 years of data were analyzed.

7. Temp Reg. § 1.482-9T.

8. The selected SIC codes included: 731 - Advertising; 732 - Consumer Credit Reporting Agencies, Mercantile Reporting Agencies, and Adjustment and Collection Agencies; 733 - Mailing, Reproduction, Commercial Art and Photography, and Stenographic Services; 736 - Personnel Supply Services; 811 - Legal Services; 872 - Accounting, Auditing, and Bookkeeping Services; 8740 - Management and Public Relations Services; 8741 - Management Services; and 8742 - Management Consulting Services.

9. See Treas. Reg. §§ 1.482-1 (d)(1) and 1.482-5(c)(2), and Temp. Reg. § 1 .482- 9T(d)(3) (factors of comparability).

10. Return on total operating cost (ROTCC) = Operating Income/Total Operating Costs. ROTOC is synonymous with the term "markup" that is used throughout this article.

11. Rate-of-return analyses have been successfully used in the tax courts concerning transfer pricing matters where the preference was noted for a large, statistically significant analysis, rather than a limited number of comparables. See, for example, the expert testimony provided by Dr. Irv Plotkin in the both the DuPoitt and the Limited Stores, Inc. cases. £./. du Pont de Nemours & Co. v. United States, 79-2 U.S.T.C. «1 9633, adopting 78- 1 U.S.T.C. 1.9374, 608 F.2d 445 (Ct. Cl. 1979), cert, denied, 100 S. Ct. 1648 (1980); In re Express, Inc., Nos.812330-34 (N. Y. Div. Tax App. 1995).

12. The All Marketing Codes group includes both marketing companies that perform routine and non-routine functions. Again, no companies were eliminated from the SIC codes within this group. The performance of this group over time stresses the importance of evaluating each company to ensure they are comparable to the routine marketing functions being performed to qualify under the SCM and an at-cost charge.

AUTHOR_AFFILIATION

Christopher Desmond is a Managing Director and Wes Cornwell is a Director of Ceteris, Inc.'s Chicago office. Mr. Desmond advises clients on a variety of transfer pricing matters, including expert witness support, as well as global and state documentation strategies. He received a B.S. degree in Management and an M.B.A. degree with a concentration in International Business from Eastern Illinois University, and previously worked for PricewaterhouseCoopers LLP. Mr. Cornwell advises clients' senior management on a variety of transfer pricing, economic, and valuation issues. He obtained his B.A. degree in Economics with a concentration in Applied Statistics from the University of Michigan, and is pursuing an M.B.A. degree at the University of Chicago, and formerly worked for PricewaterhouseCoopers LLP. The authors express their appreciation to Mike Heimen, Ryan Lange, and lohn Wiora of Ceteris, Inc. for their assistance in the preparation of this article. The authors may be reached at Christopher.Desmond@ceterisgroup.com and wes.cornwellô'ceterisgroup.com.