International Co-operation and Tax Administration Division
Centre for Tax Policy and Administration
Organisation for Economic Co-operation and Development

In this comment letter, we will first comment on the group ratio rule, and then, also present our view on the fixed ratio rule which is supplemented by the group ratio rule. It is crucial that the BEPS recommendations be consistently implemented under the inclusive framework, and that any remaining issues, including the design and operation of the group ratio rule, be steadily worked on. Keidanren expects the OECD to continue to play the leading role in these endeavors.

1. Group Ratio Rule

(1) Group's Net Third Party Interest Expense

Of the proposed methods of calculating a group's net third party interest expense, which serves as the numerator of the group ratio, Approach 1 should be adopted that uses interest figures in the consolidated income statement without adjustment. Given that global companies have a few hundred to more than a thousand subsidiaries, Approaches 2 and 3 are far from feasible in that the former requires adjusting interest figures in a group's consolidated income statement, and the latter demands identifying and measuring necessary items of interest and payments economically equivalent to interest.

Paragraph 9 of the Public Discussion Draft points out that the downside to Approach 1 is the risk of "manipulation" of interest figures in the consolidated income statement. However, considering the fact that consolidated financial statements normally go through rigorous accounting audits, it is highly unlikely that there is a substantial risk of companies resorting to such manipulation for the purpose of inflating the numerator of the group ratio.

Paragraph 9 also states to the effect that differing definitions of interest income and expense resulting from the adoption of different accounting standards or policies might decrease fairness among corporate groups. We believe, though, that further consideration is needed as to whether such differences are "substantial" enough to justify sacrificing the simplicity of the system.

Attention must also be paid to the fact that, even if the country of domicile of a group's ultimate parent adopts Approach 1, its effect would be negated should Approach 2 or Approach 3 be adopted by the country of domicile of its subsidiary. This is because a subsidiary itself does not possess the data needed to conduct the calculations required by Approach 2 or Approach 3; thus there is no other option but for the ultimate parent to perform calculations in accordance with the laws and ordinances of the country of domicile of the subsidiary. For that reason, any country introducing the group ratio rule should adopt Approach 1 and respect the accounting standard used by the ultimate parent company.

Additionally, in order not to increase the complexity of the system, it is desirable to minimize room for each country's adjustments to the definitions of interest income and expense that reflect its tax policy goal. From that perspective, some potential approaches suggested in the Public Discussion Draft warrant careful consideration: in particular, an approach to exclude net interest expense paid to a related party with 25% or greater control from the definition of net third party interest expense; and that to adjust a group's net third party interest expense so as to include the group's share of the net third party interest expense of its equity-accounted associates (paragraphs 27 and 30).

(2) Group-EBITDA

Calculations of group-EBITDA, which serves as the denominator of the group ratio, need to be as simple as possible, as with those of the numerator. Non-recurring items should in general be included in group-EBITDA without adjustment, as recommended in paragraph 64 of the Public Discussion Draft.

A problem is that the existence in a group of a subsidiary with negative EBITDA would correspondingly reduce group-EBITDA, resulting in the unreasonably high group ratio. Moreover, if a group as a whole posts negative EBITDA, the group ratio rule no longer works as a system supplementing the fixed ratio rule.

To address the first problem, we consider it reasonable to introduce a certain cap, as a rule. As for the second problem, we disagree with excluding a subsidiary with negative EBITDA from the calculation of group-EBITDA because such exclusion would increase the administrative burden.

In view of these deficiencies the group ratio rule has, what is important is not to address the deficiencies within the boundary of the group ratio rule alone, but for each country to set its fixed ratio sufficiently high and establish an effective carryforward system under the fixed ratio rule. It is essential that, by doing so, the impact on the majority of companies not involved in BEPS be minimized and a greater administrative burden be prevented.

2. Fixed Ratio Rule

The fixed ratio rule lies at the heart of the BEPS final report on Action 4. Although this rule is anticipated to help, to a certain degree, prevent BEPS involving excessive interest expense deductions, we still have some concerns as expressed below.

The first one is the definition of interest subject to the restriction. Because the fixed ratio rule is recommended to be applied to all of a group's net interest expense, including that paid to lenders within the same country and to third parties, even interest expense on loans from domestic banks would be subjected to the restriction. Developments like this would hamper ordinary business operations, undermining incentives for investment. Further, with interest income being taxable in the recipient entity, double taxation might arise in a large scale.

Whereas the BEPS final report on Action 4 states that BEPS can occur if interest is paid to a domestic lender who makes a corresponding payment to a foreign lender or if interest is paid to a third party under a structured arrangement, we believe that such cases are very exceptional. If a jurisdiction determines that the BEPS risk posed by interest deduction is not material, then the jurisdiction should adopt the more targeted fixed ratio rule which is applied to net interest expense paid to foreign group entities.

The second concern is the definition of EBITDA. The fixed ratio rule dictates that nontaxable dividend income be excluded from EBITDA so as to prevent debt from being used to fund tax-exempt income. Yet, nontaxable dividend income is, in almost all cases, simply the result of business investment, not the goal. Therefore, when incorporating the fixed ratio rule into domestic law, each country should be given flexibility to determine the appropriateness of excluding any nontaxable dividend income from EBITDA.

For example, assume that Company P located in country X borrows money in country X to finance its large-scale research and development activities in country X and to fund equity investment in its subsidiary, Company S, which operate as a manufacturer in Country Y. Company P may be disallowed to deduct its interest expense to some extent under the fixed ratio rule in country X, since Company P's EBITDA can be relatively small because of the huge R&D costs it bears and nontaxable dividend paid by Company S. Moreover, the group ratio rule in country X may not give relief to company P, since group-EBITDA can be relatively large due to the large depreciation costs incurred by Company S's manufacturing activities. In this case, although company P does not have any intention to cause BEPS, disallowed interest expense would arise. Such a situation needs to be avoided.

The third concern is the level of a fixed ratio. While it is recommended that each country set its fixed ratio within a range of 10% to 30%, there is a risk that, if group-EBITDA has declined as a result of deteriorating economic conditions, disallowed interest expense may arise. Even though the fixed ratio rule permits the carryforward of disallowed interest expense, each country should apply a fixed ratio that is high enough to minimize the impact on ordinary companies.

It is true that a portion of the interest expense disallowed under the fixed ratio rule may be deducted under the group ratio rule. Nonetheless, as made clear once again in this Public Discussion Draft, the adoption of the group ratio rule is likely to dramatically increase the administrative burden, depending on its design. We are gravely concerned that each country may build a system whereby a fixed ratio is set so low as to force numerous companies to rely on the group ratio rule. A system should be designed that is completed with the fixed ratio rule alone to the extent possible. Even if there is a case where a taxpayer needs to proceed to the group ratio rule, it should be as simple as possible.

Sincerely,

Subcommittee on Taxation
KEIDANREN