January 25, 2023
Tax Treaties, Transfer Pricing and Financial Transactions Division
Centre for Tax Policy and Administration
Organisation for Economic Co-operation and Development
By email to email@example.com
Re: Business Roundtable comments on OECD public consultation on “Pillar One – Amount B”
Business Roundtable welcomes the OECD’s commitment to working multilaterally and with the private sector to ensure sound tax policies and straightforward tax administration, which are essential to protecting investment and economic growth.
On behalf of more than 230 chief executive officers of America's leading companies, Business Roundtable is pleased to submit comments in response to the OECD’s public consultation document of December 6, 2022 on Pillar One – Amount B.
Amount B should:
- be simple to apply and administer for taxpayers and tax authorities and should minimize controversy;
- be broader in scope (including retail, digital, cloud and other services, and entities that do more than just baseline marketing and distribution activities) and, as much as possible, use scoping criteria, such as de minimis thresholds, based on objective quantitative measures that can be computed from benchmark data;
- apply to all Covered Groups subject to the Amount A rules;
- be based on economic analysis consistent with the arm’s length standard;
- not provide optionality to countries or permit the use of local comparables;
- not impose onerous documentation requirements (such as some of those proposed in the consultation document), and;
- be used as the Marketing and Distribution Profits Safe Harbour threshold instead of the convoluted and complex MDSH proposed by the Secretariat. If a Covered Group is already reporting taxable profits from marketing and distribution activities in a jurisdiction (including via withholding taxes) in excess of Amount B, such excess should reduce Amount A for that jurisdiction.
Methodology for physical goods and digital goods and whether any adjustments would need to be made
In our experience, the TNMM can be applied to the sale of digital goods and services where either a buy-sell or commissionaire structure is set up. The functions performed for sales of digitally delivered products are similar to those performed for tangible goods; the difference is simply in the delivery. As long as the distributor’s functions conform to the baseline functions prescribed by the Amount B rules, distributors of digital products should be able to use those rules.
We note that the tech industry commonly uses distributors of hardware, software and components to benchmark companies that distribute digital products. Certain working capital adjustments can be appropriate to account for differences in inventory and payment terms. Functional intensity adjustments (e.g., SGA/sales) may also be appropriate to allow additional return for relatively greater sales activity. Allowing for these may be a way to address the Inclusive Framework’s concerns about ancillary activities performed by the distributor.
Since the objective of Amount B is to simplify benchmarking, and adjustments add complexity, any adjustment mechanism should be easy to implement, be clearly explained and rely on data that is readily accessible to the taxpayers and tax authorities.
Qualitative vs. quantitative factors - is the balance right? If not all factors are met, would it be appropriate to make an adjustment for those factors?
As a general comment, we believe both the scoping approach to Amount B as well as the related development of the comparable benchmarking analysis should be based, as much as possible, upon quantitative measures. Such an approach could reduce controversy regarding the baseline marketing and distribution activities within scope of Amount B. Certain qualitative characteristics could also be considered for scoping, but such criteria should be clearly defined.
With respect to the search criteria described in Annex A, the database filtering and qualitative factors listed in the annex are fairly standard. The qualitative review is based on the business descriptions contained in the database, which is usually pretty brief. In practice, economists often need to review websites or annual reports to get enough qualitative information.
The use of keywords to identify non-comparable activities, while helpful in screening a large volume of potential comparables, may lead to rejecting viable companies when the keyword may be used in a different context, such as describing customers’ industries. In addition, the use of just the database business descriptions to identify companies with other activities may not produce accurate results due to limited information in the database.
Another approach to consider is to use only quantitative screens, which would remove the subjectivity and greatly simplify the process but increase the number of observations. However, since Amount B will be based on a point within the interquartile range and the additional companies will likely include companies at the high and low ends of the profit spectrum, the overall impact to the interquartile range may not be significant. It would be interesting to compare the results of this approach to the results of the proposed Amount B rules once the technical analysis is complete. If the ranges are close, a strictly quantitative approach could be considered for simplicity.
While there is value in the regression analysis work being done in terms of identifying industry or geographic differences and relationships between key variables, using the output in the Amount B price setting will increase complexity. The pricing matrix and mechanical pricing tool described in Section 4.2 of the consultation document seem complicated and prone to subjectivity which could increase disputes.
Whether there should be exceptions to the rules, and if so, what are the benefits of creating those exceptions?
These comments relate to the exemptions and exclusions section of the consultation document.
If a reliable comparable uncontrolled price (CUP) exists for a transaction that would otherwise be subject to the Amount B rules, then the taxpayer should be allowed to use the CUP. Use of a CUP can reduce disputes if the CUP is truly comparable (preferably an internal CUP), and the taxpayer should be able to demonstrate why it is preferred over the TNMM. We note, however, that truly reliable CUPs in the context of some distribution structures are hard to find. If such a CUP is identified (e.g., if an internal CUP exists or for certain commodities transactions) there should be clear guidelines on comparability to ensure that it meets the required threshold.
The proposed local market comparables exception should be discarded as it could complicate, rather than simplify, the benchmarking process and will be a source of potential controversy. Many jurisdictions do not have reliable publicly available information (e.g., Africa and parts of APAC). Based upon our audit experiences, we are concerned that local benchmarks have been impacted by opportunistic inclusion of comparables, some of which may not be publicly available or do not share a functional profile consistent with baseline marketing and distribution activities. Moreover, it is time consuming to conduct numerous local comparables searches and seems wasteful when the local result is not very different from the regional result. Should the exception for local market comparables be allowed, this will go against the stated purpose of Amount B to streamline the benchmarking process.
The proposed non-tangible goods exclusion should also be reconsidered. As noted above, to the extent the software/digitally delivered product distribution functions are similar to traditional distribution functions conducted by a buy-sell entity or commissionaire, reliable benchmarking can be done to calculate an Amount B. As the Inclusive Framework further develops the methodology, they can consider whether certain adjustments would be appropriate to account for differences in distributing tangible versus intangible goods. Excluding non-tangible goods would prevent large tech companies that are subject to Amount A from benefiting from the Amount B simplification. In our view, if a company has to comply with the Amount A rules, then it should be able to use the Amount B rules as well.
We also consider that the Amount B scoping criteria could reliably incorporate retail distributors as well as wholesale distributors. The functions, assets and risks of retail distribution are not dissimilar to wholesale distribution and the consultation does not put forth any reasoning for why there should be disparate treatment. To the extent the Inclusive Framework members perceive a difference between wholesale distributors and retail distributors, any specific functional differences that may be a cause for concern should be identified so industry members are able address those concerns directly and determine whether there are approaches to mitigate such concerns, if applicable.
The activities exclusion also needs to be reconsidered. In practice and especially after the Covid pandemic with increasing work location flexibility, distributors/sales entities may have employees performing additional functions such as R&D. Moreover, specialized services are commonly performed by distributors/sales entities in the tech industry (e.g., maintenance contracts, technical services). Again, it is not practical to segment these functions into a separate legal entity.
While it is undoubtedly cleaner to have separate legal entities for R&D, manufacturing, etc., it is not always practical or cost effective for an MNE to set up multiple entities in a jurisdiction and therefore entities with multiple business steams under separate contractual arrangements are common. In such cases, taxpayers have processes to segment the financial results of legal entities and if a taxpayer can demonstrate a reliable and systemic approach to segmentation, the marketing and distribution activities should remain in scope. One possible approach, especially where segmentation may not be easily accomplished, would be to use a de minimis threshold for other activities in the Amount B rules. Functional intensity adjustments could also help address this as we would expect the tested party to have additional operating expense for these services.
Pricing model comments
While Return on Sales is the most commonly used profit level indicator, the Berry ratio may be an appropriate alternative for certain industries or distributors with high turnover. The Berry ratio cap-and-collar is an additional mechanism to ensure Amount B results fall within the expected results of independent companies performing comparable functions. Additional guidance on how the Inclusive Framework members would propose to operationalize the Berry ratio will be useful, including any views on the segmentation of cost data into COGS and OPEX for purposes of applying such a net profit indicator. The applicability of the Berry ratio can be further evaluated once the benchmarking results are shared.
It is critical that the Amount B pricing methodology is aligned with the arm’s length principle. The Amount B pricing methodology must be supported by robust and publicly available benchmark data that reflects a broad spectrum of marketing and distribution activities across industries and geographies. The pricing analysis should be updated regularly and consistent with the suggestion in the Consultation Document should include a multi-year range of data (e.g., 6 to 8 years). Regarding the range, we believe that, in practice, targeting a point or narrow range may be difficult or require year-end adjustments.
Regarding adjustments, we suggest that the Inclusive Framework consider working capital adjustments to reflect differences particularly in inventory levels and potentially applying adjustments to account for higher tested party functionality (e.g., SG&A/sales) if required to make tax authorities comfortable with broadening the scope of Amount B to include more distribution/sales entities. However, the application of any adjustments must be clearly defined for ease of implementation and consistency and to reduce areas of dispute.
Regarding financial data, we believe that it would be overly burdensome to require that multi-year financial data for non-taxpayer tested parties and detailed segmentation by customers be provided upon request. This is especially true if the taxpayer transacts with multiple tested parties.
In addition, the written contract requirement seems burdensome to the extent it goes beyond the existing intercompany agreement. Paragraph 87(k) of the consultation document does note the taxpayer can supplement agreements for terms not explicitly covered. Additional clarity on the form of this supplemental information is requested (e.g., an appendix in the local file).
Need for further consultation
We note that the draft provisions in the consultation document have not been agreed upon by the Inclusive Framework, and a significant number of issues are likely to be the subject of debate within the Inclusive Framework in the coming months. The public should be given the opportunity to provide comments on a future draft of the Amount B rules after open issues have been discussed within the Inclusive Framework.
Business Roundtable urges the Inclusive Framework to take the above comments into account in its work on the Pillar One, Amount B rules. We appreciate your consideration of these comments. Please do not hesitate to contact us if you have any questions.
Vice President, Tax and Fiscal Policy
+ 1 202-467-5266