INSIGHT: Understanding IRS’s New Transfer Pricing Functional Cost Diagnostic Model

On March 1, 2019, the IRS Advance Pricing and Mutual Agreement (APMA) program announced the introduction of a new spreadsheet model, referred to as the functional cost diagnostic (FCD) model. Accompanying the model, was a memorandum simply titled, Functional Cost Diagnostic Model. The memorandum summarizes the intended purpose of the model and provides instructions on how the model is to be used, including the steps necessary to complete the model. The model is intended to be prepared by taxpayers that are seeking an advance pricing agreement (APA).

Why Was the FCD Model Released?

According to the APMA, the FCD model was developed to “facilitate its review of the taxpayer’s APA request … APMA believes it is necessary to consider whether the arm’s-length values of the respective contributions to the proposed covered transactions might be more reliably measured by comparing them to one another than by benchmarking returns for functions as a single taxpayer in the proposed covered group performs, the assets it employs, and the risk it assumes.” (APMA, Functional Diagnostic Model, Section 1, Feb. 15, 2019.).

APMA’s latest annual APA report from March 2019 indicates that the CPM/TNMM, a one-sided test, was used for 86% of APAs executed in 2018. The FCD model, however, is based on the two-sided residual profit split method (RPSM) as described under Treasury Regulation Section 1.482-6. The RPSM first identifies the subject business profits as a whole. These are next attributed to the parties based on their “routine” (or “benchmarkable”) contributions. The “residual” profit, which is attributable to non-routine (non-benchmarkable) contributions or intangibles, are then divided based on the capitalized and amortized costs associated with the development and maintenance of those non-routine contributions and assets. Of course, if only one party is deemed to have non-routine contributions, the RPSM will reduce to a one side method because all the residual profit will flow to the entity making those contributions.

The release of the FCD model seems to be in response to growing use of (or reference to) the results of profit split models in an increasingly complex transfer pricing environment within companies and between taxing authorities. At minimum, it may be used as a sanity check on the results of ultimately agreed methods. As such, the FCD model is likely to be used by APMA and taxpayers in APAs for two purposes:

1. As a resource in applying the residual RPSM;

2. As a comparative tool for the potential results from using other methods (such as a one-sided CPM/TNMM analysis);

The approach clearly indicates one way the IRS may internally frame and interpret a taxpayer’s facts and the proposed transfer pricing method in APA cases. Especially of note is its consistency in looking at historical investments and returns on assets which underlies much of the IRS’s and Treasury’s historical view of transfer pricing. This can be seen in in the prior analytical spreadsheet tool APMA required taxpayers to complete, as well as the application of non asset-based profit level indicators within Treas. Regs. Section 1.482 including operating asset adjustments to returns under the comparable profits method.

It appears this information for a RPSM was being used or referenced so regularly by other tax authorities, that it warranted standardization and a request across all APA cases. Statistics on methods used as the primary method in APAs will provide insight on whether use of the RPSM has increased, though it will likely take several years before any trends on this point can be discerned. In addition to its use in APA cases, the IRS have also indicated that the FCD model may also be used or referenced in general transfer pricing examinations outside of the APMA program.

Workflow and Key Inputs to the FCD Model

The FCD model instructs taxpayers to identify and list historical and projected functional costs to (i) contributions that are routine in nature, and (ii) contributions that are non-routine in nature or that develop and support non-routine intangible assets. The costs associated with routine functions are treated as same period expense, while the cost and expenses associated with non-routine contributions are capitalized and amortized as economic assets, of which the amortization is treated as a period expense.

The profits to routine contributions are deducted from the overall business profits in the same period, but the economic asset balance associated with non-routine contributions is used to share the residual profits between the parties which effectively provides a multi-period return to those contributions.

The chart below outlines the elements of the model using the example provided by APMA, while key inputs to the model are outlined below the chart.

(a) The beginning of the model outlines the historical and future projected income statement and balance sheet for the parties in consolidation. The model also requires the process of segmenting financials by functional activity, which will likely require assumptions and allocations that are not readily available by the taxpayer. The inputs to the FCD model outline the number of historical and projected years of financials that will be entered. Historical costs are necessary to determine each parties’ share of non-routine capitalized investments and should at least be equal to the sum of the estimated lead time and useful life of the functional costs being analyzed.

(b) Functional costs that are routine in nature and can be reliably benchmarked are identified. The model links these costs directly from the financials. Taxpayers are then instructed to enter the profit level indicator (PLI) and return that is attributable to each routine activity. Sufficient support for the benchmarking assumptions should be referenced and kept on file. Note, there are no options for inputting ranges. As such, one data point must be selected.

(c) Non-routine functional costs by party and type are identified and listed. Assumptions on the estimated lead time (e.g., the time between the date of an investment and date economics benefits are realized) and the estimated economic useful life (the period that an investment generates returns or losses) will require the taxpayer to receive guidance from operations teams that support lead time and useful life assumptions. The model automatically capitalizes these costs based on the selected inputs using standard formulas and techniques.

(d) The residual profits (total profits less routine—or benchmarked—profits) are then allocated based on each parties’ contribution share calculated from the prior step.

(e) Finally, each parties’ targeted operating profit will be equal to the sum of profits earned for its routine activities and its share of residual profits. The FCD model illustrates this as an adjustment, which is the difference between each parties’ targeted (calculated) operating profit per the analysis less reported operating profit.

Initial Takeaways and Key Considerations for Using the Model

Based on our review of the FCD model, we make the following observations and initial areas to consider related to its use:

• In order for the FCD model to be internally consistent, historical costs and expenses for all non-routine contributions are required over their full asset lives; otherwise the non-routine economic asset balances will be incomplete.
• The FCD model and its results are dependent on the assessment and delineation of routine and non-routine (benchmarkable vs. non-benchmarkable) contributions and activities, and their related costs and expenses; these are subjective determinations that may take more time to agree in more complex transfer pricing environments such as networks, evolving business models, or highly centralized control of local activities and contributions.
• By focusing on costs as the measure of key contributions, the FCD model assumes that all costs for all contribution types are equally contributive to non-routine profits, and as a result, may downplay qualitative contribution differences, or be distorted where there are high levels of “routine” or external costs that mask the real underlying costs that better reflect relative non-routine contributions.
• The capitalization of the non-routine costs and expenses in the FCD model is performed on a “nominal value” basis (i.e., based on booked historical costs), as opposed to “real value” basis, which may not adequately take into account the difference in risks associated with the costs and expense and their timing.
• The FCD model does not include functionality for calculating each parties’ contributions to non-routine activities outside of using relative capitalized and amortized costs. Use of actual costs or third-party market references to allocate non-routine contributions are not an option in the model, even though Treas. Reg. Section 1.482-6(c)(3)(i)(B) outline these as potential methods to allocate residual profits.
• We note that the FCD model is set up for the assessment of “on-going” transfer pricing and not intangible asset valuations, which are now complicated by the inclusion of goodwill and going concern in the list of compensable intangibles.
• Care and thought will be also needed in its use and construction where a company has been acquisitive or undergone a reorganization or restructuring that captures the value of future profits into current taxes in one jurisdiction of another.

While the FCD model is quite prescriptive, it’s clearly a useful start point for a transfer pricing analysis in more complex situations. While there can be pitfalls to its use, many of these can be mitigated through basic best practices and relatively simple reliability adjustments so that issues can be narrowed and resolved to taxpayer specific facts and circumstances. These best practices should include basing the analysis on a thorough and considered evaluation of the business, its critical success factors, and where those contributions are managed and controlled within the bona fide legal framework of the arrangements. These should reconcile with the proposed transfer pricing practices, and public information (investor reports, presentations, and SEC filings).

Taxpayers using the model proactively should maintain sufficient support for their decisions on the input assumptions and consider performing sensitivity analyses for changes in these inputs and the impact on results to ensure the allocation of risks and rewards remains consistent with the underlying arrangements and activities.

With the FCD model still being early in its mandatory use or reference, it remains to be seen what implications it will have on the outcome of APA’s or the methods used under the APMA program, IRS transfer pricing examinations, and transfer pricing in general. However, it is clear that when the RPSM is used the IRS is focused on making sure that historical investments for which deductions have been made in US tax returns are a consideration in the evaluation of transfer pricing.

A copy of the FCD model and guidance memorandum can be requested via email to lbi.ttpo.apma.feedback@irs.gov. You should include “FCD Model Request” on the subject line.

This column does not necessarily reflect the opinion of The Bureau of National Affairs, Inc. or its owners.

Author Information

Rod Koborsi is a director and Simon Webber is a managing director at Duff & Phelps LLC in Silicon Valley.