Limiting Base Erosion

S. 3921. Introduction

The different Actions of the BEPS Project deal with various problems using different mechanisms through setting minimum standards, addressing best practice recommendations and providing interpretative guidance. The diversity of the proposed actions aims at achieving a coherent and holistic approach to limit base erosion and profit shifting where the shortcomings of applying a particular regime should be captured and addressed by the rules of another Action. Although the direction is undoubtedly set towards a converging reflection of economic substance, it is inevitable that the proposed mechanisms would remain conceptually different and evolve at a different pace and with different objectives.

The hybrid mismatch actions focus mainly on neutralizing tax-abusive outcomes and offer little in defeating the causes, namely the misalignment between the debt and equity classification criteria across the different OECD jurisdictions. On the other hand, in an attempt to compensate for the sound impact on the world of corporate finance, the interest barrier rules of Action 4 keep it technically simple by relying substantially on financial reporting definitions and concepts. Less attached to the roots of civil law and statutory accounting, the transfer pricing actions certainly go farthest in the search for economic reality by revising the arm’s length standard and designing an approach for delineation of the transactional characteristics according to their genuine business substance.

This contribution examines the interaction of Action 4 of the BEPS project: Limiting Base Erosion Involving Interest Deductions and Other Financial Payments with the other Actions dealing with hybrid mismatch arrangements and transfer pricing outcomes. The author first attempts to explore the sequence of application of the different rules by establishing a hierarchy between the fixed ratio of Action 4, the primary and defensive rules of Action 2 and any transfer pricing adjustments that may in parallel delineate a funding return interacting with the scope of the former. The contribution then takes a closer look at some of the hybrid mismatch examples provided in the appendices of Action 2 and elaborates on the implications that arise from applying the measures of the other Actions.

2. Application of the “interest barrier” along with the hybrid instrument rules

2.1. Hierarchy in which the rules apply

The OECD points out that states should implement rules for neutralizing hybrid mismatches alongside their general interest deduction limitation regimes. The S. 393reason is that general restrictions such as interest barrier or thin capitalization rules still provide room for claiming double deduction or deduction/non-inclusion outcomes when hybrid financial instruments are being used. However, in the author’s view, since most interest limitation regimes are designed to allow a certain amount of debt capacity, the reclassification of a debt arrangement as an instrument of equity under Action 2 needs to be factored in for the purposes of the general tests as well, to avoid double counting.

For instance, State A allows deduction of interest expenses up to 30% of the taxpayer’s stand-alone EBITDA irrespective of the status of the creditor and the terms of the financing arrangement. However, it allows “escaping” from the interest barrier if the taxpayer demonstrates that its debt-to-equity ratio is practically identical with that of the rest of the group, e.g. it is not abnormally leveraged beyond the group capital structure. Under the assumed fact pattern, the taxpayer is financed partially with a hybrid debt instrument which gives rise to a tax deduction in State A but no-inclusion in the state of the instrument holder, State B. If State A disallows interest deduction on the instrument in line with Action 2, then it seems reasonable and proportionate to exclude the value of that same instrument from the capitalization tests when applying the escape clause of the general interest barrier rule. Not doing so would prevent deduction of other interest expense on regular debt instruments due to the hybrid loan still being regarded as liability, thus consuming much from the stand-alone debt capacity of the taxpayer.

Action 2 does not explicitly provide a recommendation in this respect but acknowledges the potential problem and suggests that “the interaction between the interest limitation rule and the hybrid mismatch rules should be co-ordinated under domestic law to achieve an overall outcome that avoids double taxation and is proportionate on an after-tax basis”. Along that line, the OECD recommends in both Actions 2 and 4 that the rules to address hybrid mismatches should always be applied first before the fixed ratio is tested so the latter can take into account all net payments treated as interest for tax purposes.

The above order of application is also important with a view to facilitating the effective functioning of the defensive rule under Action 2 in situations where the source state does not apply the primary rule for disallowing interest deduction. If the aforementioned example is considered again but now it is assumed that State A does not recharacterize the nature of the hybrid debt arrangement and treats the interest accruing on it no differently, then the state of the creditor (or as may be seen in State B – the shareholder) should have to apply the defensive rule for S. 394taxing the corresponding income. This should need to be applied irrespective of the fact that the interest expense may be disallowed on general grounds under the fixed ratio in State A. Therefore, it is not only the outcome that matters, but also the grounds on which deduction is denied in the state of source. Any other reason for restricting interest deduction, such as interest barrier or thin capitalization limitations, loss-making activity or arm’s length adjustments seem to be insufficient to prevent inclusion of the hybrid income in the taxable base of the instrument holder.

2.2. What is ordinary income?

What is even more interesting in the case at hand is whether the state of the holder would regard the included income as recharacterized interest income or as taxable dividend, i.e. simply withdrawing its exemption in compliance with the defensive rule for D/NI outcomes. The classification of the income type is critical with respect to the holder’s own debt shield (assuming its state operates interest barrier rules as well). If the taxed income is regarded as interest, it would at least allow equivalent room for deductible interest expense including potential unwinding of interest brought forward from previous periods. That would not be the case if the income is viewed as non-exempt dividend, since such a type of income would only contribute to the holder’s tax adjusted EBITDA, hence allowing just 30% headroom for interest deduction.

The issue here is that, in fact, no recharacterization of the instrument takes place or at least no such recharacterization is intended under Action 2. The OECD points out that the primary and defensive rules are limited to adjusting the tax consequences that flow from the difference in the tax treatment of the instrument and should not generally affect the underlying character of the payment (e.g. whether it is treated as interest or a dividend). Instead of trying to align the treatment of the financial instrument in the affected jurisdictions, Action 2 only ensures that no unfair advantage can be taken from the mismatch, thus discouraging the use of hybrid instruments for international tax arbitrage.

The defensive rule requires that the exempt income generated on the hybrid instrument is integrated in the tax base of the holder as ordinary income. Ordinary income is defined in the text as a payment under a financial instrument generally including interest, dividends and other investment returns that are subject to tax S. 395at the payee’s full marginal rate. Therefore, recharacterization of the type of income is left to the discretion of each jurisdiction. Although one could draw a conclusion that coordination is necessary, stemming from the remark discussed above that there has to be achieved an overall outcome that avoids double taxation and is proportionate on an after-tax basis, no specific recommendation is made in this respect by the OECD.

If a state employs special classification criteria on recognition of debt and equity for tax purposes and in line with them designates a financial instrument held by its taxpayer as an equity investment, it may find it inconsistent to recharacterize the instrument as debt by importing the treatment of the state of the issuer. This may effectively be the case each time when the corresponding jurisdiction is not compliant with the OECD primary rule and the state of the holder needs to activate a defensive rule. The said state may not be willing to reclassify the equity instrument into debt not only for the sake of consistency and legal certainty but to prevent an adverse impact on its corporate tax collection. Otherwise, the imported classification of the equity return as interest income on debt may allow the instrument holder to increase its interest deduction capacity and/or claim a deduction for impairment losses on debt receivables.

The issue of different classification has been to an extent resolved under EU law with the proposed text of the Anti-Tax Avoidance Directive (ATAD) according to which the mismatch must always be neutralized through a denied deduction at the state of source. Simply put, since Member States are all bound to the same primary rule, there is no need to impose a defensive linking rule. However, the proposal for amending the ATAD with a view to extending the EU anti-hybrid rules to arrangements with third states includes a defensive linking rule enforcing Member States to include any income on hybrid financial instrument which has given rise to deduction in a third state. Therefore, since the proposal does not elaborate further on the classification of the included income as either dividend or interest, the issue described above is yet to be resolved with respect to hybrid mismatches arising with third states.

2.3. Hybrid returns on the list

Although it is a matter of domestic law to determine whether the included ordinary income should fall within the interest or dividend category, the states that implement interest barrier rules, as recommended in Action 4, may have to consider whether the return on the particular instrument would also fall within the list of payments that are economically equivalent to interest. Action 4 provides a S. 396non-exhaustive list of financial payments that are recharacterized as interest for the purposes of the fixed ratio test. As the interest barrier rule tests the amount of net interest expense to EBITDA, then any ordinary income regarded as interest income would in favor of the taxpayer reduce the amount of net interest expense subject to restrictions. It is clear that when a type of payment is regarded as interest, it should be treated as such no matter whether it gives rise to income or expense.

The list starts with a few items that traditionally fall within the concept of a hybrid instrument, such as payments under profit-participating loans; imputed interest on instruments such as convertible bonds and zero coupon bonds; amounts under alternative financing arrangements, such as Islamic finance. Nevertheless, a certain number of jurisdictions which impose their own tax criteria on classification of debt are likely to treat any forms of loans that accrue interest in direct correlation with the profitability of the issuer as equity financing. Therefore, the return on such profit- participation loans or bonds with mandatory conversion clauses would likely be regarded as dividend income for domestic law purposes. In the author’s view, if this is applied to the case discussed above, the state applying the defensive rule and reintegrating the previously exempt dividend in the tax base as ordinary income would need to allow the taxpayer holding the instrument to offset its interest expense with the reclassified dividend for the purposes of the fixed ratio. Of course, this is valid only if the particular state does not specifically exclude the return on certain instruments such as profit-participating loans from its definition of economically equivalent payments. The latter is, however, harder to do prospectively in the Member States of the European Union as the Anti-Tax Avoidance Directive proposal explicitly includes the list provided in Action 4 under its own definition of borrowing costs.

The Directive further elaborates that the “exceeding borrowing costs” computation should factor in any other economically equivalent taxable revenues. However, to the extent that the Directive relies on the primary rule to disallow deduction at the level of the issuer, it does not preclude the corresponding Member State from granting exemption for the funding return earned by the holder of the instrument. Therefore, even if that return fits into the definition of an economically equivalent revenue (e.g. payment on a profit-participating loan), it would still not be able to reduce the “exceeding borrowing costs” as long as it fails to meet the “taxable” requirement. As pointed out in the preceding paragraphs, the situation could be different when the mismatch is resolved by the application of the defensive rule and the return is treated as taxable ordinary income.

S. 3973. Interest-free financing across borders – as complicated as it gets

3.1. Focus on actual payments

The next section of the contribution will explore the classic problem of interest-free financing. Albeit the simplicity of its terms, the non-interest bearing loan contract seems to be nowhere near to simple, straightforward treatment post-BEPS.

Although Action 2 focuses on the hybrid outcome of an arrangement, it makes it clear in a number of paragraphs that the scope of the recommended rules extend only to outcomes arising from actual payments on financial instruments. The definition of “payment” in the text specifically excludes payments that are only deemed to be made for tax purposes and that do not involve the creation of any new economic rights between the parties. This triggers a number of issues when it comes to interest-free debt instruments that may still give rise to interest income and expense in line with the concepts of financial reporting and transfer pricing.

Action 2 illustrates the different implications that two arrangements with very similar fact pattern have depending on the distinctive approach to imputed and deemed or notional interest (terms used quite interchangeably in practice, which, however, will be given different meaning for the purposes of this section of the contribution).

3.2. Imputed discount on non-interest-bearing debt

Companies that prepare their accounts under IFRS or certain other established accounting frameworks are often required to adjust the value of particular non-current liabilities by discounting them to their present value. The rationale behind this is to reflect the time value of money and factor in the income statement for the particular period the extra cost that the entity effectively bears to be able to defer the settlement of its liability to a future moment in time. This is typical in situations where a subsidiary receives a substantial interest-free shareholder loan to start its operations without the burden of paying interest in the first periods characterized with lower liquidity. As independent creditors would unlikely provide funding to a regular business borrower free of any charge, it is apparent that the lending is extended by the creditor in its capacity of a shareholder and this has to be reflected for financial reporting purposes.

The principal due at the maturity date of the loan is regarded as the future value of the liability, which is discounted to reflect the present value of the debt. The S. 398difference between the two values is effectively deemed as an informal capital contribution by the shareholder. Over the life of the loan contract, the subsidiary accrues interest expense that is accumulated in the discounted present value of the debt so at the maturity date it reaches the full amount of principal under the contract. Then, it is deemed that the subsidiary settles its interest payable to the parent company by virtue of repaying the entire principal.

In jurisdictions where tax accounting follows to a great extent the accruals-based concept of IFRS or similar comprehensive framework, the subsidiary would likely be able to get a tax deduction for the imputed interest expense no differently than any other cost of borrowing. Conversely, if the parent company in the above example is based in a jurisdiction that has its own tax accounting rules that allow interest-free shareholder loans or impose tax on contracted interest on a receipt basis, it is likely that no taxable event will ever take place.

Action 2 views the above situation as a hybrid mismatch within the scope of the primary and defensive rules. The key question here is not whether the instrument gives rise to a deduction/non-inclusion (D/NI) outcome, which is apparent, but rather if there is any actual payment on the instrument. The OECD has taken the position that in this particular case the imputed interest arises in respect of a repayment obligation and therefore it is regarded as a D/NI payment on a hybrid instrument.

Once determined that the imputed interest expense should be disallowed for tax purposes under the primary rule, it has to be established whether the loan would still come into play in the interest barrier tests. The Action 4 list of economic equivalent payments does not specifically include imputed interest on non-interest-bearing loans. And in fact, despite its relatively common use in intercompany funding structures, the author finds it unusual that Action 4 barely addresses the issue. The said list, however, explicitly includes imputed interest on convertible and zero-coupon bonds and one can argue that the latter carries substantial similarities to interest-free loans, and hence the treatment should not be too different.

Further guidance could be traced in the later sections of the report where Action 4 stipulates that when calculating the net interest expense to third parties for the S. 399purposes of the group ratio, the group can remove from the test any notional interest amounts which do not include actual payments of interest as these are not economically equivalent payments. Since Action 4 does not deal separately with what constitutes a payment, the author finds it appropriate to stay consistent with the interpretation given by the OECD in the Action 2 example discussed above. As a result, it could reasonably be argued that the imputed interest expense constitutes a payment on a financial instrument and as such it should generally take part in the computation of the fixed ratio under Action 4.

3.3. Notional interest recognized for tax purposes only

Following the example reviewed above in respect of the imputation of interest on a shareholder loan, Action 2 uses another example to highlight the conceptual difference that applies to situations where an actual payment cannot be separately identified. In the particular case, the subsidiary does not discount its liability and presents no imputed interest expense, and so does the lender. Nevertheless, the state of the subsidiary allows notional interest deduction on the shareholder loan for tax purposes. The example concludes straight to the point that since no payment takes place, no hybrid mismatch arises.

The author finds this treatment inconsistent and misaligned with the objectives of the recommended measures of Action 2. In both examples reviewed, the subsidiary has the same contractual obligation to repay the same amount of funds borrowed from the parent company. The difference is that in the first scenario the subsidiary presented part of it as interest expense in line with the applicable accounting framework, whereas in the second case the interest was deducted for domestic tax purposes only. The rationale behind this different treatment is presumably that the imputed interest arises in respect of a repayment obligation and therefore it cannot be left out. It is not clear to the author, though, why the notional interest deduction allowed by the subsidiary state in the second scenario should not be linked to the fact that there is a financial liability (that is such because of the obligation to deliver cash or another asset to a creditor). Had there been no liability, the state would have unlikely allowed interest deduction on it. Notwithstanding the above, the OECD concludes that the interest deduction arises in respect of an amount that is not capable of being paid. Accordingly, there is no payment under the financial instrument that gives rise to a D/NI outcome. It then needs to be explored further if the deducted notional interest could qualify as enough of a “payment” for the purposes of the fixed ratio under Action 4.

S. 400The example given in Action 2 does not elaborate on the type of the domestic law regime that allows notional interest deduction, e.g., a special domestic rule incentivizing shareholder funding or a transfer pricing adjustment. Action 4 leaves out of its scope notional interest deductions allowed under domestic law and points out that further work will be done by the OECD addressing these cases. However, the text specifically refers to items not being treated as economically equivalent to interest when they arise as a result of applying a specified percentage to the equity capital of an entity. Therefore, the case at hand should not fall in this specific category as long as it is interest on debt and not on equity.

The most common reason for attributing interest deduction to an interest-free loan from a related party is the arm’s length principle laid down in the transfer pricing rules and recommendations. As obtaining funds for no charge is apparently not a market-based transaction, the parties that have entered into it are supposed to adjust their tax base as if the loan were at arm’s length. Such a transfer pricing downwards adjustment seems to be out of the scope of the primary rule even if the lender makes no corresponding upwards inclusion in its tax base. As analyzed earlier, the reason for this is that the tax adjustment is done for tax purposes only and no actual transfer of value, i.e. payment, takes place. Following that position, it could be reasonably concluded that the transfer pricing adjustment could not qualify as a payment economically equivalent to interest either, based on this no-payment criteria.

The list provided in Action 4 regards as interest any items measured by reference to a funding return in line with the revised transfer pricing guidelines (this contribution addresses this in the following sections) but these again involve actual value transfers, for example, from a high-tax jurisdiction to “cash-box” entities based in low tax havens. Then, if the adjustment is not a qualifying payment economically equivalent to interest, it has to qualify as per se interest in order to be factored in the fixed ratio.

Therefore, states implementing the recommendation of Actions 2 and 4 should, in the author’s view, provide clear additional domestic measures to prevent disproportionate deduction outcomes arising from cross-border interest-free financing. In that respect the OECD highlights in Action 4 that further work is planned to be done with respect to transfer pricing on financial transactions as the related aspects impact considerably the rules on interest deduction limitation.

S. 4014. Mismatches in the context of lease instruments

4.1. A classic lease hybrid

Another area where hybrid financing, transfer pricing and interest limitation rules interact significantly is the use of lease instruments in a group context. Leases traditionally have been regarded in two categories: financial and operating, where finance leases are economically treated as debt-financed acquisition of assets and operating leases are simply viewed as asset rental arrangements. The forms of the agreement could vary significantly giving rise to different legal relationships between the lessor and the lessee, but the economic substance of the arrangements normally comes down to either debt financing or temporary renting. In that sense, transactions such as a sale of asset and its finance leaseback are recharacterized under IFRS and certain other accounting frameworks as providing of a loan accompanied with a legal transfer of the title of an asset to secure the borrowing, while the seller/lessee retains economic ownership. Notwithstanding the above, the classification of leases as finance or operating is often not a clear-cut exercise and there is considerable room for discretion and interpretation.

Action 2 explores an example where two related entities from different jurisdictions have entered into a sale and leaseback arrangement. Company A of State A sells an asset to its related buyer Company B in State B for which A receives sales proceeds. A then, in need of the asset, leases it back from B under a lease arrangement. State B applies a substance-over-form approach and treats the sale and leaseback as a loan extended from B to A in return for which A would be paying principal installments and interest. A, however, treats the transactions as a one-off sale followed by a rental arrangement in line with the form of their contractual relationship. As a result, B receives interest income and return of principal while A incurs expenses for rental charges. Since the return of principal is not taxable income, B pays tax only on the amount of accrued interest income, whereas A deducts the entire rental installment, taking advantage of the tax treatment in State A following the legal form.

According to the text of Action 2, the hybrid financing rules would not apply in State A because State A sees no financial instrument but only a rental contract. Therefore, even if there is a mismatch due to a different classification of arrangements, from the perspective of State A such a mismatch is out of scope of the rules S. 402for payments on hybrid financial instruments. State B sees the arrangement as provision of debt and hence it theoretically qualifies for the defensive rule. Nevertheless, it fully taxes the interest income earned on the debt whereas the rest is designated as return of principal. Action 2 points out that where the counterparty does not treat the payments under the arrangement as payments under a financial instrument, the hybrid financial instrument rule should only apply to the extent of the funding return. In other words, provided that the return of principal is not an exempt equity return, i.e. dividend, it falls outside the scope of the hybrid instruments rules as well.

Uncaptured by the rules for hybrid mismatches, it needs to be established whether the arrangement would be limited by the interest barrier rules of Action 4.

State B sees interest income flowing into Company B under a financial instrument (debt receivable). Therefore, the realized income should definitely be factored in the reduction of net interest expense for the purposes of the fixed ratio, either as per se interest income or as the finance cost element of finance lease – an item specifically enlisted as a payment economically equivalent to interest.

At the same time, State A sees rental payments on the temporary use of assets. The rules on hybrid instruments of Action 2 have to be applied in order of priority first but as it has been determined, the recharacterization of the arrangements is out of their scope and so is the D/NI outcome in that case due to not originating from a financial instrument. Action 4 then explicitly stipulates that the best practice approach should not apply to operating lease payments and takes the latter out of the definition list of payments economically equivalent to interest. Therefore, Company A would be able to deduct the entire rental expense without being restricted under the interest barrier rule to a percentage of A’s EBIDTA. In the same time Company B could benefit from the interest income qualification to increase its deductible leverage capacity.

In the author’s view, the above examples illustrate that in certain fact patterns the Actions may fail to interact in an effective manner to achieve a holistic and comprehensive approach to countering international tax arbitrage. Actions 2 and 4 do not attempt to align the treatment provided under the domestic laws of the affected jurisdictions, but instead apply objective tests to counter the targeted outcomes. In other words, the Actions seem to rely on discouraging BEPS practices by limiting the effect rather than the cause of it, i.e. the mismatch in classification of identical transactions under the laws of different states. This may be a practical S. 403approach as the opposite, i.e., harmonizing the tax treatment of financial transactions across multiple jurisdictions, might hardly be possible in reality. Therefore, in the author’s view, the BEPS project seems to be leaving a whole lot to be dealt with by the unbiased and inherently international field of transfer pricing.

4.2. Interaction with the revised arm’s length principle

Taking the lease example above, it could be further assumed that State B is a low-tax jurisdiction and Company B is a capital-rich entity that performs treasury and other financing functions within the multinational group. For that reason, A has obtained proceeds from B through the sale of a specific asset and now is leasing it back to employ it in its uninterrupted business activity. It is further assumed that Company B has no expertise in managing, maintaining, replacing or insuring this type of assets and there is no active market for rentals of such specialized and customized equipment. As pointed out in the fact pattern above, Company A pays rental charges to B for the use of the said asset.

According to the revised guidance for applying the arm’s length principle, State A should identify the economically relevant characteristics of the relationship between the related parties with a view to accurately delineating the actual transactions taking place. The revised guidelines stipulate that where the characteristics of the transaction that are economically significant are inconsistent with the written contract, then the actual transaction should be delineated in accordance with the characteristics of the transaction reflected in the conduct of the parties. If Company A is economically exploiting the equipment before and after the sale and B is in reality providing funding without being capable of carrying out lease activities, especially in the context of temporary renting to a base of customers, then Company B may not be entitled to more than a funding return.

This would mean that Company A’s rent expenses should be effectively reduced for tax purposes to interest cost accruing on the financing arrangement. Such funding cost may be even lower than the one recognized by Company B as interest income in its financial statements if the latter is not able to demonstrate its treasury and financial management capacity to provide financing.

Once Company A’s rental transaction is replaced through a transfer pricing adjustment with the cost of funding return, the interest barrier rules are supposed to kick in again. This time the Action 4 list explicitly classifies as payments economically equivalent to interest any amounts measured by reference to a funding S. 404return under transfer pricing rules. Therefore, the expensed funding return should be subject to tax deduction according to the fixed ratio testing of Company A’s EBITDA. The rest of the installment may either be reclassified as a loan principal repayment or simply derecognized for transfer pricing purposes.

In such a way, the transfer pricing rules close the gap with economic substance that may be left open by tax accounting due to not following the substance-over-form principles of IFRS or simply IFRS allowing accounting policy choices and other interpretation leeway. This is reflected in the Action text as follows:

5. Delineating a funding return for transfer pricing purposes

This contribution will finally examine another situation where transfer pricing adjustments may be required to capture finance income left outside the scope of the interest barrier rules of Action 4. This is the example of a trading company carrying out working capital-intensive activities and incurring substantial interest costs with no corresponding income in that same finance category. This case is not separately examined in the Actions but the OECD refers to its implications when guiding how net interest expense to third parties should be computed for the purposes of the group ratio.

The fact pattern could be illustrated as follows – a multinational group uses a specialized trading entity that distributes products manufactured by other group companies on given markets. The function of the trading company is to purchase goods and then resell them to independent retail chain stores. The manufacturing process performed by the group is labor-intensive and is characterized by considerable payroll costs that are incurred on a monthly basis. At the same time, the specifics of the market in which the trading company distributes and competes are such that retail stores exercise pricing pressure and push for generous credit terms. This places the trading company in a difficult intermediary position where it has to pay immediately or in advance for the acquisition of inventory from the group manufacturing unit, so the latter could afford to meet its regular payroll obligations and supplier payables. On the other end, the trading company needs to offer extensive sales credit terms where receivables are usually settled not earlier than a few months. As a result, the trading company faces a considerable workS. 405ing capital gap for the period between the dates when its payables are due and its receivables are collected. To fund this gap, the trading company would normally need to borrow from a bank and incur substantial interest expense. The income statement of the trading company would then show operating profits from the sale of goods equal to the prevailing market margin for distributors, reflecting the working capital bridge functions. The operating profits would be accordingly reduced sizably by interest expense on the bank loan. Since the trading company’s state operates an interest barrier rule, the deduction of its interest expense is limited to 30% of the company’s EBITDA.

The above examples show how disproportionately the general interest limitation rule could apply when it relies on the presentation of the type of income for financial reporting purposes. The role of the trading company apparently comes down to two core functions in the group value chain, i.e. distribution of products and working capital financing. Since the sales income is presented as gross revenue from sale of goods, it does not allow netting off with the corresponding interest expense incurred on the bank loan to fund the sales. The issue can normally be resolved by the principles of financial reporting, and specifically IFRS, which prescribe that sales revenue should be measured at fair value of the consideration received and thus, when the cash inflow is deferred, the fair value of the consideration would be less than the nominal amount of the cash received or receivable, whereas the rest would be considered to be imputed interest income. The interpretative literature on the topic states that when an arrangement effectively constitutes a financing arrangement, the fair value of the consideration would reflect market participants’ assumptions about the time value of money and risk associated with the financing arrangement. While this may often be applied to strip the financing element from a sale transaction with substantial deferred payment terms, according to the author’s observations, it is not widely used when the credit period is short-term and relatively standard for the particular market, or not used at all if the entity is reporting under a less sophisticated accounting framework.

Therefore, assuming that the trading entity from the example above could not rely on the presentation of interest income for financial reporting purposes, it may be adversely impacted by the thresholds of the interest barrier rules as recommended under Action 4. Looking again at the list of payments economically equivalent to interest, one cannot find any similar type of income being reclassified as interest for the purposes of the fixed ratio. Moreover, it may be difficult to sustain a position that part of the sales revenue earned from the trading transactions is a return received in relation to the raising of finance by the retail stores (although many could reasonably argue along that line).

S. 406Nevertheless, when discussing the group ratio clause, Action 4 recommends that there may be different approaches to computation of the interest expense figure and if a state decides to base the test on adjusted financial reporting figures, it may consider the addition of interest income or expense recognized within a different category of income or expense. The text specifically stipulates that this could include interest income that is included within gross revenue, or interest expense that is included within the cost of sales or in the tax line. In some cases, these amounts may not be identified in a group’s consolidated financial statements, and will need to be obtained from underlying financial information. Groups may be able to introduce processes to identify these payments more easily, in particular where this would mean an increase in total net third-party interest expense. Again, in the author’s view, if this approach is followed for group ratio purposes, there is no reason why a particular group entity should not use it as well for stand-alone testing. If this is the case, clearly, the trading entity in the example above would be able to offset its bank interest expense with the delineated funding return and suffer no or little disallowed costs under the interest barrier regime.

The OECD has long established that for the purposes of better comparability, a transfer pricing analysis should factor in any working capital adjustments in an attempt to adjust for the differences in the time value of money between the tested party and potential comparables, with an assumption that the difference should be reflected in profits. The question is whether such an adjustment could also be used to recharacterize the type of the income as a separate item from the tested company’s operating profits.

The revised guidance on the arm’s length principle stipulates that the transaction as accurately delineated may be disregarded, and if appropriate, replaced by an alternative transaction, where the arrangements made in relation to the transaction, viewed in their totality, differ from those which would have been adopted by independent enterprises. The context in which the revised guidelines are developed imply that the exercise of an accurate delineation should be performed when there is perceived non-compliance with the arm’s length principle due to differences between the economic characteristics of the deal and its contractually arranged allocation of risks and functions. This could be the case, for instance, if the trading company earns a distributor’s margin that compensates risks related to the collection of receipts, insurance over products, regulatory compliance, but in fact it does not have the capability to manage the contractually assumed risks if they crystalize. According to the fact pattern illustrated above, the trading company may in that case be entitled to not more than a (risk-free) funding return for bridge financing the customer credit terms.

S. 407In such a case the funding return should, similarly to the lease case addressed earlier in the contribution, qualify as a payment economically equivalent to interest. As a result, the trading company would be able to net off its bank interest expense with the delineated funding return when applying the fixed ratio. The question is how this would have been dealt with if the distributor’s margin was at arm’s length and no reduction to a limited scope of financing activities was necessary, i.e. no transfer pricing adjustment was made at all. In the author’s view, states implementing both the interest barrier rules and adhering to the OECD’s recommendations on transfer pricing should facilitate a mechanism where a taxpayer could voluntarily delineate its transactions according to their business substance for the purposes of placing them in the finance/non-finance category. In that respect, the “raising of finance” criterion may be fair to encompass not only long-term financial instruments but any other debt arrangements that can be reliably measured by reference to the time value of money.

6. Conclusion

The OECD Actions have taken a big step towards limiting the effectiveness of international tax arbitrage through the use of debt financial instruments.

Still, much has been left to the domestic laws of each jurisdiction for classifying what is debt and what is equity and whether an arrangement could be designated as a financial instrument in the first place. This classification serves as a starting point for application of the hybrid mismatch rules. Nevertheless, the set of rules proposed by Action 2 is not intended to break the hybrid tie by aligning the form of the arrangement with its economic substance and place it in the appropriate category, but rather to try to counter an outcome that may take advantage of the system mismatch. This offers not much of a helping hand to the interest barrier rules of Action 4, meant to be applied alongside and compliment the limitation of excessive leverage by testing financial ratios, since these ratios are based again on the original domestic classification of the instruments and the types of return they generate. This is especially valid with respect to the classification of returns integrated as “ordinary income” by applying the defensive rule under Action 2.

The concept of economically equivalent interest of Action 4 goes a step further in that direction as it manages to capture substitute items that represent the cost of finance albeit being accommodated in other forms. Nevertheless, an attempt to apply economic substance over the legal form while at the same time providing an exhaustive list of arrangements covered by the scope of Action 4 offers no solution to classic examples of arbitrage such as interest-free loans, lease classification mismatches and presentation of credit sales revenue. Inevitably, the time value of money is comprehensively embedded in various business models and supply S. 408chains that extend beyond the traditional forms of raising debt finance. Therefore, limiting its impact based on designated categories of income and instruments (especially left to the discretion of diverse domestic tax law and accounting rules) would hardly be able to achieve a completely proportionate and effective international tax system post-BEPS.