On 17 February 2015 a public consultation was held on Action 4 of the OECD/G20 action plan on base erosion and profit shifting (BEPS). This action aims to limit base erosion and profit shifting that takes place through interest and other financial payments. The OECD will develop recommendations on dealing with the use of related party debt to obtain excessive interest deductions, including the use of financial payments that are economically equivalent to interest.

Discussion draft

On 18 December 2014 the OECD published a discussion draft on interest deductions and other financial payments and invited interested parties to submit their comments. Interest payments on related party debt can reduce the profit of an entity in a particular country even though the whole group may have little or no external debt. Typically related party interest payments would be paid from a high tax to a lower tax jurisdiction, reducing the overall group liability to tax.

The OECD Working Party has looked at the best practices currently employed by various countries to discourage profit shifting by the use of related party debt. The OECD discussion draft set out options for the types of measure that could be used to prevent tax avoidance through financial payments. The draft considered what is interest and what are payments equivalent to interest, and who the legislation should apply to. It also looked at the question of whether the new rules should apply to the level of debt or to the interest expense itself, and whether the gross or net position should be considered.

The draft looked at whether interest payments should be limited by reference to the position of the group to which the entity belongs by comparing the entity position to the group-wide position; or whether interest should be limited by reference to a fixed ratio, such as a maximum percentage of interest to EBITDA; or perhaps a combined approach could be taken.

The discussion draft also considered the role of targeted rules; a small entity threshold; the treatment of non-deductible interest expense and double taxation; considerations for groups in specific sectors; and interaction with other areas of the BEPS action plan.

There will be interaction between Action 4 and the parts of the BEPS action plan concerned with transfer pricing, hybrid mismatch arrangements, controlled foreign companies and prevention of treaty abuse.

Points raised during discussions

Questions were circulated for consideration at the conference on 17 February 2015. There were delegates from individual country tax administrations, the large advisory firms, industry bodies such as the CBI, multinational groups and the OECD’s business and advisory council (BIAC).

Arm’s length principle

Best practice should be based on the arm’s length principle (ALP); however the OECD is considering taking these measures in addition to the ALP. Very few countries apply the ALP to any extent to restrict interest deductions and they generally do not use the ALP as a tool to tackle BEPS. There is a focus in the discussion draft on the level of debt held in various countries, and this cannot be dealt with using the ALP. Also, the location of debt in a high or low tax jurisdiction is something that the ALP cannot necessarily deal with. So the focus of Action 4 is not on the ALP but on alternatives.

The work on deductible payments is at the heart of the BEPS project, together with the transfer pricing issues. But debt is crucial to the funding of a business and the measures taken to combat BEPS must be targeted to ensure that they affect only BEPS related activities without affecting normal commercial activity. There are existing best practices in place in various countries that can be considered.

Statistics

The OECD statistics division noted that a lot of interest is paid by MNEs in countries with higher statutory tax rates. Interest shows up more in higher tax rate countries and the affiliates in those countries often have higher interest to EBITDA ratios. PWC has compiled statistics on behalf of BIAC and these were presented to the meeting. These statistics showed the proportions of the 100 largest MNEs that would fail a test based on interest to EBITDA if the limit was 10%, 20% or 30%. Among small cap companies the proportion of companies failing the test would be higher.

The largest 100 MNEs are not necessarily representative of all MNEs. The 100 largest groups can borrow at lower interest rates, have robust cash flow and are not so reliant on debt, so their debt levels tend to be lower than in smaller companies. Some even have net interest income rather than interest expense. Also companies in some sectors, such as those with a lot of physical capital, have higher levels of debt, for example those in the utilities sector.

The criteria for the fixed ratio test cannot therefore be based on the 100 largest companies. A test based on these companies would catch many small cap companies and capital intensive companies, not just the outliers. The statistics used for the study presented at the conference are based on third party interest only, as they are taken from consolidated financial statements, so related party interest does not show up.

Interest rates are volatile. A test that is based on current rates would not be suitable if rates change – the interest to EBITDA ratio would change through no fault of the MNE. An informal study made of the available statistics revealed no correlation between interest rates and EBITDA so ratios would change as the interest rates went up and down in each country.

Group-wide interest allocation rule

The group-wide test has the advantage that it takes the third party debt of the group and the interest paid on this, and then allocates the net third party interest expense to the jurisdictions in which the group is operating on the basis of its economic activity in those jurisdictions. The third party interest should be tax deductible and the whole point of the exercise is to give a tax deduction in the group just for the third party interest, in the jurisdictions where the group’s economic activity takes place.

When calculating the group wide allocation of debt it is practical to only take into account the group itself as presented in the financial statements. If an attempt is made to take into account the full results of entities that are not fully consolidated there are complex considerations to take into account. There are large numbers of joint ventures (JVs) and a group may own a majority but there are restrictions on putting debt into the JV. These JVs are common in developing countries with local partners and the restrictions should be taken into account otherwise MNEs with many JVs may be unfairly penalized.

The OECD paper on hybrid entities and instruments has a 25% rule for entities to be taken into account and that paper has a lot of attribution rules which can be very complex. The complexity comes from attributing ownership through chains of entities and looking at who owns what and what debt should be brought into the computation.

Where an entity is not included in the consolidated financial statements there would be a reason for this – materiality; cost to gather data too high. If there is another option to leave it out under the law of a particular country one would need to look at the details of the opt-out.

Basis for measuring economic activity in a country

Debt should be related to economic activity in a particular country but how do you measure economic activity – by assets or earnings? This decision of assets or earnings could be taken by each MNE based on its activities, provided that it is applied consistently rather than being cherry picked from year to year.

Both measures of economic activity raise significant issues. Earnings do not necessarily reflect the need for capital e.g. a service business needs less capital than a manufacturer. Assets are rather more related to borrowing but fair market valuation would be a burden for business and for the tax administration. Book value is known but is not the right measure of economic activity. Book value undervalues self-created intangibles. Fair market value for internally generated intangibles is impossible to establish. For certain industries e.g. property-related industries book value might be permissible so the MNE could choose what is best for it.

Limitations of self help

A group could help its own interest deductibility position under group-wide allocation rules by moving debt into countries in proportion to its economic activity in those countries. This self-help cannot always work however due to measures such as: currency controls, capital adequacy rules, capital reduction rules, buy-back rules, business purpose tests for tax relief; and specific restrictions in industries such as oil and gas. There are other restrictions in smaller countries. So it is impractical to push debt down completely in proportion to economic activity. Also to maintain deductibility the MNE groups would need to vary the debt levels in each country year by year and forecast where debt will need to be in a particular year.

There are always operational issues – dividend restrictions, foreign exchange provisions, and these mean that debt is more flexible than dividends. MNEs use debt because it is flexible – interest deductibility restrictions could therefore reduce investment on a global basis.

Fixed ratio test

A fixed ratio test could look at the level of an entity’s interest expense or debt. It could take into account gross interest expense or net interest expense (i.e. interest expense less interest received). The delegates tended to favor the use of net interest expense. Much of the discussion focused on a fixed ratio based on a percentage of interest to EBITDA. Delegates tended to favor higher ratios to ensure that normal commercial operations were not subject to an interest disallowance. A ratio of 30% was frequently referred to.

Combined approach

The discussion draft presents two options for a combined approach. Approach 1 would have the group-wide interest allocation rule as the general rule, with a carve-out from the general rule for entities that meet a low fixed ratio test; and approach 2 would have a fixed ratio rule as the general rule with a carve-out from the general rule for entities which meet a group ratio test.

The delegates discussed the advantages and disadvantages of including a combined approach in the recommendation, and the strengths and weaknesses of combined approaches 1 and 2 in the discussion draft. In the view of a number of delegates option 2 on the discussion draft is preferred – a mix of fixed ratio tests or a back-up of a group-wide allocation test (with the proviso that the latter can be complex).

The opinion was expressed by some delegates that combined approaches could have benefits, as in option 2 in the draft, but the fixed ratio rule must be designed to apply widely across most groups. The group-wide rule should only be a backup for a fixed ratio test with a fixed ratio that is sufficiently high.

Situations where a targeted rule would be required

Third party interest rules are important – the objective should be to achieve a deduction for third party interest expense – but some of the tight general rules will disallow some of that third party interest expense. In the view of some delegates there is therefore a need to be a bit more generous with the general rules and then to have targeted rules to make sure that profit shifting is specifically targeted. The key aim is to allow a deduction for third party interest expense.

These interest deductibility rules are the most complex set of rules in many countries. There is already a sophisticated combined approach in many countries e.g. in the UK with a combination of arm’s length principle, General Anti Abuse Rule (GAAR), anti-arbitrage rules, the debt cap/worldwide gearing tests and thin capitalization approaches. These rules are targeted so any general rules suggested by the OECD should be flexible.

Other issues

The delegates considered the possibility of a carry forward of disallowed interest expense or of unused capacity. This could address issues of volatility in the application of a rule or mismatches where interest expense and earnings/ assets arise in different periods.

They also considered how could an interest limitation rule could be designed that would address BEPS risks posted by banks and insurance companies, while not having an undue impact on the regulatory position of a group. The special considerations for other businesses such as the property sector and the oil and gas sector were also discussed.