Preparing for Transfer Pricing Audits: Bayer’s Battle with the Canadian Revenue Agency

The following article is part of a series of case studies and articles which have been kindly contributed by Dr Harold McClure, a New-York based economist who has advised in the area of transfer pricing for many years.

Action 13 of the Base Erosion and Profit Shifting project requires large multinationals to prepare a Master File along with Country-by-Country (CbC) reporting in addition to local documentation. Before the adoption of these requirements, many multinationals prepared separate documentation reports for each of their key local affiliates. One concern has always been whether such local documentation represented sufficient information of the transfer pricing for the local affiliate

The audit of Bayer’s Canadian affiliate by the Canada Revenue Agency (CRA) is a cautionary tale of what multinationals should consider as additional information beyond the typical local documentation report if they face scrutiny of their transfer pricing by a foreign local tax authority.

Bayer is not a client of mine so any information noted in this discussion comes from its Annual Reports or the July 20, 2020 court ruling in Bayer Inc. v. the Attorney General, which considered the scope of information that Bayer must provide to CRA. The court decision noted that the CRA began a transfer pricing audit of Bayer Canada for the 2013 to 2015 period, which focused on the transfer prices paid by Bayer Canada for pharmaceutical products.

A Simple Application of the Transactional Net Margin Method (TNMM)

Transfer pricing audits often begin with a request of the taxpayer’s documentation report. While Bayer may have produced an informative and detailed transfer pricing analysis for the relevant issues involving its Canadian affiliate, many multinationals simply produce a very high level TNMM analysis with the local affiliate as the tested party. Bayer’s sales are approximately $50 billion per year and if the Canadian market represents 4 percent of these sales, Canadian sales represent $4 billion per year.

Bayer's 2015 Annual Report also notes that the consolidated operating margin was nearly 13 percent for the audit period. Let’s further assume the following:

• Canadian selling expenses = 30 percent of sales;
• Production and R&D expenses borne by the parent = 57 percent of sales.

Consolidated profits on sales to Canadian customers are $260 million per year, which are allocated between the parent and the Canadian affiliate depending on the transfer pricing policy. The following table presents the income statement for the Canadian affiliate and the parent corporation on sales to Canadian customers under the assumption that the Canadian affiliate receives a 35 percent gross margin.

Hypothetical Income Statement for Bayer Canada

The standard TNMM analysis asserts that the parent corporation owns all intangible assets including product, process, and marketing intangibles. This assertion is the justification for arguing that if the distribution affiliate receives a routine return for selling activities, then the intercompany pricing must be arm’s length. A 5 percent operating margin should be seen as a reasonable estimate of this routine return even if transfer pricing practitioners struggle with finding third party distributors in the life science sector that are functionally comparable to a high function distributor of branded pharmaceuticals. Wholesale distributors of generic pharmaceuticals tend to have gross margins near 5 percent as their operating expenses are near 3 percent of sales. While their operating margins are low, their return on operating expenses are very high.

The Data Requests from the CRA

Selecting appropriate comparable companies and the choice of the profit level indicator for a TNMM analysis, however, might not be the most serious problem with the transfer pricing documentation we suggested. The court decision noted a very broad information request from the CRA. After a series of requests issued in 2018, the August 21, 2018 query number 17 requested:

Pursuant to our discussion on July 18, 2018, we would like to audit agreements made between any member of the Bayer Group with third party(s) in force during the 2013 and 2014 taxation years that perform some or all of the following activities in regards to pharmaceutical products:

• Are located in an Organization for Economic Cooperation and Development (“OECD”) member nation;
• Perform research and development (clinical trial level stage II, III and/or IV, inclusive);
• Perform regulatory compliance activities (notice of compliance, product labelling verification, etc.);
• Perform client and corporate product support;
• Perform quality control and assurance activities;
• Regional marketing and sales activities (e.g. detailing, medical affairs);
• Chain supply management activities (e.g. purchasing. distribution);
• Price negotiations with local regulatory bodies;
• Price negotiations with the public (e.g. provincial formularies) and private (e.g. insurance companies) funding bodies.

Please provide a matrix of no less than 50 contracts that meet some or all of the criterion [sic] listed above, and make sure activities contemplated in the agreements are highlighted, so that CRA can select contracts for further review.

On November 14, 2018 the CRA issued requests for foreign based information. The representatives objected to these data requests. While the court decision noted that the CRA was entitled to request relevant information, the court order limited the scope of these requests to the agreements with the 21 named pharmaceutical and life sciences companies that operate at arm’s length from Bayer. The tone of these data requests appear to be inquiries as to whether Bayer Canada was engaged in more than distribution activities, which could raise issues with respect to the ownership of intangible assets.

Master File Requirements

A well prepared Master File could address many of the concerns of the local tax authority. The broad categories of the information to be provided in a Master File include:

• The multinational’s overall business and the key drivers of consolidated profits;
• The multinational’s key intangible assets and which legal entity owns each intangible asset;
• The multinational’s overall financial and tax position; and
• The multinational’s intercompany financing activities.

The detailed information includes a listing of the top markets, a description of the main geographic markets for those products, and a functional analysis describing the principle contributions to value creation by each affiliate. The detailed information also includes a list of the key intangible assets, listing of the key R&D facilities and how R&D is managed, and agreements with respect to cost sharing and contract R&D arrangement as well as license agreements.

The information in the CbC report indicates how worldwide profits are allocated across the various affiliates but does not directly provide guidance as the intercompany policies that led to the actual allocation of worldwide profits. The head count and functional headcount information required in the CbC report, however, could prove useful.

Note that the CRA initially wanted to know what R&D activities were conducted by Bayer Canada. If this head count information showed that all of its employees were only involved in either selling or marketing activities, then this issue becomes moot. Of course, it is possible that phase III trials on some of Bayer’s branded products were conducted by Canadian employees then a question with respect to the ownership of product intangibles could be raised. If the parent compensated the Canadian affiliate at fully loaded costs plus a reasonable markup in a contract R&D arrangement, the assertion that the parent owns the product intangibles is warranted.

The Annual Reports for Bayer provide segmented financial information for its four business lines:

• Branded pharmaceuticals;
• Consumer health;
• Crop services; and
• Covestro (polyurethane products).

The branded pharmaceutical segment is more profitable than the other three lines of business and represents approximately 30 percent of sales.

The focus of CRA’s transfer pricing audit is this branded pharmaceutical segment. An aggregate TNMM approach would not necessarily address the transfer pricing issues without segmentation of the financials for the Canadian affiliate.

Roche Australia and Speculation Where This Audit May Be Heading

What information is needed to evaluate any particular transfer pricing issue in part depends on the transfer pricing approach. This CRA audit of Bayer Canada reminds me of the issues in Roche Products Pty Ltd v. Federal Commissioner of Taxation.[1] Roche is a Swiss parent that sold various products to Australian customers via its Australian sales and marketing affiliate, Roche Products Pty Limited. These products included branded pharmaceuticals, reagents and other diagnostic products. An expert witness for the Australian Tax Office (ATO) segmented the Australian affiliate’s financials by line of business and used separate TNMM analyzes for each line of business.

Over the 1993 to 2002 period, the operating expenses for the pharmaceutical division of the Australian affiliate totaled 39 percent of sales but gross profits represented only 37 percent of sales. The TNMM analysis proposed separate markups for selling expenses and for marketing expenses, which suggested operating profits should be 5.4 percent of sales. This analysis suggested that the overall gross margin for the pharmaceutical division should be 44.4 percent.

The expert witness for Roche argued that the Resale Price Method was the preferred approach. This expert also argued actual sales fell short of expected sales as an explanation why the TNMM approach would overstate the arm’s length gross margin.

Even though Roche sold several branded pharmaceuticals, this expert focused on one third party agreement where the third party distributor received a 37 percent gross margin. Another expert witness for the ATO, however, noted that the gross margins for other third party agreements for Roche products exceeded 40 percent. While the Resale Price Method is a viable approach for the Bayer Canada branded pharmaceutical transfer pricing, the question remains which if any of the 21 third party agreements requested by the CRA represents a comparable third party agreement.

A dynamic profits based approach may be considered if no truly comparable transaction can be found for the application of the Resale Price Method. TNMM ignores the dynamics of what is essentially a market share strategy, which is described by section 1.482-1(d)(4) of the US transfer pricing regulations.

In certain circumstances, taxpayers may adopt strategies to enter new markets or to increase a product’s share of an existing market (market share strategy). Such a strategy would be reflected by temporarily increased market development expenses or resale prices that are temporarily lower than the prices charged for comparable products in the same market. Whether or not the strategy is reflected in the transfer price depends on which party to the controlled transaction bears the costs of the pricing strategy.

Market share strategies under arm’s length pricing often witness periods where the upfront market expenses are very high relative to early sales such that overall operating expenses exceed gross profits. As sales grow, the operating expense to sales ratio declines below the gross profit margin, so the local affiliate enjoys sufficient long-run profits to compensate it for both its routine return as well as a reasonable return on its initial investment by incurring the upfront marketing. A reliable application of this approach requires understanding what expected sales over the life of the market share agreement, which may differ from what actual sales turned out to be.

This transfer pricing audit is ongoing so it is unclear what the CRA might argue. The public record also does not inform us as to the various facts that have been gathered by the CRA. Their information requests suggest the focus of the transfer pricing audit comes down to what is the appropriate gross margin for the Canadian affiliate on its selling and marketing activities with respect to branded pharmaceuticals. If the CRA proposes a transfer pricing adjustment based on an application of the Resale Price Method using the 21 third party agreements obtained in this litigation, the taxpayer might consider an alternative approach such as an application of the market share strategy.

[1] 2008 ATC 10-036.