BEPS Action 4: Interest Deductions and other financial payments Response by the Chartered Institute of Taxation

BEPS Action 4: Interest Deductions and other financial payments Response by the Chartered Institute of Taxation 1 Introduction 1.1 The Chartered Institute of Taxation (CIOT) is pleased to respond to the Public discussion draft published on 18 December 2014 on BEPS Action 4: Interest Deductions and other financial payments. 1.2 The discussion draft on Action 4 focusses on group-wide interest limitations based around a group s external borrowings. We are far from convinced that an interest limitation rule is the best option to address the BEPS concerns arising around interest deductions. 1.3 Although it is noted (at paragraph 5 of the discussion draft) that there is a sense that unilateral action by countries is failing to tackle some of the issues at the heart of the problem, we note that other Actions of the BEPS project are addressing a number of base erosion issues arising in respect of interest deductibility, and differences between countries regarding the treatment of debt are a natural part of an international tax system where countries retain sovereignty over tax matters. 2 Policy considerations 2.1 We support the work which the BEPS project is doing to counter arrangements which have (rightly) caused so much public concern, namely structures whose purpose is to enable some multinational enterprises (MNEs) to pay low levels of tax in jurisdictions from which they generate significant amounts of profit. However, the proposals set out in the Action 4 discussion draft go beyond action to counter the splitting of profits from the activities which generate those profits. These proposals seek to harmonise the rules which countries have for determining how much interest is tax deductible, even in situations where the amount of profit generated is commensurate with the activity performed and in situations where no tax advantage arises.

2.2 We can see merit in proposals to reduce the scope for tax to distort commercial decision-making but whilst countries retain sovereignty to set their own tax rates, reliefs etc., tax will play a part in investment decisions. It seems to us that domestic rules regarding what expenses qualify for tax relief are, as much as headline rates of tax, matters for each country to determine taking into account all aspects of its tax system (for example, its CFC regime). As such, we take the view that the approach adopted by this consultation document goes well beyond the scope of the BEPS project. 2.3 The area addressed in this discussion draft is different from the other areas addressed by the various BEPS Actions. It considers an issue which is fundamental to all businesses that is how to finance the business. This is a very different matter to tackling, say, hybrid mismatch arrangements, which are a specific tax tool that can be used by MNEs. Every group has to finance its business. 2.4 The document seems to assume that a group is in effect a single entity and any internal financing of a group is merely an allocation of capital, and the form of this allocation is irrelevant. However, this ignores the fact that groups are made up of separate legal entities, and the transactions between them do have legal and contractual reality. The document assumes a form of formulary apportionment can be applied, ignoring the fundamental concept of separate legal personality. 2.5 The proposals do not distinguish between debt used for tax planning purposes and debt simply used to finance trading and investment activity. 2.6 We accept that intra-group loans and interest deductions can be used to shift profits. However, the vast majority of intra-group debt does not result in base erosion or profit shifting. Most internal debt financing within groups is undertaken because this is the easiest and most convenient method of financing business activities. Equity financing and the payment of dividends is a much less flexible approach, and, generally, requires far more bureaucracy. 2.7 We take the view that a basic premise should be that interest should be deductible if it meets two conditions: (i) an arms-length test ie the entity could borrow the debt in question on the terms in question from a bank or other lender; and (ii) a taxation of interest test, ie the amount deducted is matched by an amount taxed in the recipient (ignoring possible timing differences). This is one of the targeted rules noted in the discussion draft. 2.8 There is no reason, for example, why a well-capitalised and cash rich group headquartered in a relatively low taxed jurisdiction should be penalised by financing a subsidiary by debt, if that subsidiary could raise the debt externally if it were an independent entity. 2.9 We are of the view there are ways of limiting the BEPS impact of debt financing whilst retaining the ability for groups to finance themselves internally in the most convenient way. For example, if a concern is that groups can use internal debt to shift profits from high tax to a low tax jurisdiction where there is little substance, CFC rules and transfer pricing rules of the type proposed in Actions 8-10 are a better way to tackle this than interest restrictions by way of a cap. P/tech/subsfinal/IT/2015 2

3 Why groups use debt 3.1 The discussion draft appears to take the view that it is equally easy for a group to finance its subsidiaries by debt or equity. We would dispute the perception that equity is an equivalent funding option. Although funding by equity may appear to work from a purely economic perspective, in reality equity is much more bureaucratic, regulated and inflexible. As a result debt is often the commercially preferred route of financing. 3.2 In reality, inter-company debt will usually be the first choice in cash and debt management; equity financing will be a long term financing base. This is because debt is much easier to create, is more flexible in terms of repayment and is integrated into a group s treasury and cash management systems. 3.3 On a day to day basis, a group of companies will be receiving funds from various sources and locations and paying amounts out in various locations which may or may not be the same as where revenue is being received. 3.4 Very few businesses will have an even cash flow over a year. Some businesses may be geared to a seasonal event, for example, harvest in agriculture, or the return to school for a publisher of textbooks. Some subscription based businesses may receive an extremely high proportion of their cash receipts in a concentrated period. There will also be spikes in expenditure, around the time of dividend payments to shareholders or bonus season. 3.5 Typically, a group treasury function will run some form of short-term inter-company lending process, which could include cash pooling to ensure funds received in one place are able to be used elsewhere. This itself creates a large number of loan relationships. These may be transferred into longer term structural loans where a business is in a net investment/cost position; and sometimes, businesses have to make long term loans of surplus funds if there are barriers to dividend or capital repatriation. From time to time, a group will look to capitalise loans or pay dividend/return capital where it is clear loans are unlikely to be repaid. However, equity events generally require more paperwork and are subject to more conditions than loan events for example, it may only be possible to pay a dividend after a set of audited accounts are produced. There may be capital duties associated with equity issues. 3.6 Groups thus prefer the flexibility of loan financing for a variety of non-tax reasons. The proposals as put are likely to penalise groups by increasing their tax rates if they cannot constantly turn loans into equity, or eliminate upstream loans by way of dividends. 3.7 We would also suggest that affordable debt is a good thing for businesses and global growth generally. Debt indicates growth within a business and the ability to service debt should be viewed as an indicator of a strong, growing business. Debt is often a preferred route for financing a business as it can be used for growth promoting investment without diluting the interests of equity holders. 4 A group-wide interest limitation 4.1 For the reasons set out above we do not think that the G20/OECD should seek to harmonise global interest deductibility rules using an interest cap. P/tech/subsfinal/IT/2015 3

4.2 We will now detail our specific concerns around the group-wide interest limitation proposed in the discussion draft. 4.3 First, as a matter of principle, a group-wide interest limitation would be a departure from the arm s length principle and would be a significant move in the direction of formulary apportionment. 4.4 The discussion draft says that the arm s length test should not form part of this consultation process (paragraph 21). We would like to comment, however, on the concerns cited regarding the arm s length tests in paragraph 22. We suggest that, given the increasing focus, as a result of the BEPS project, on transfer pricing and reporting to substantiate all types of intra-group arrangements, most MNEs will be producing annual transfer pricing and country by country reports. Hence the expense and burden of applying the arm s length tests are already with MNEs. We suggest that these reports should be considered by countries as providing a robust safety net for tax administrations and for taxpayers. Application of the arm s length tests using these mechanisms should be seen as a valuable tool in the protection against BEPS, rather than as an irrelevance. 4.5 A group-wide interest limitation would also result in distortive effects for groups considering similar investments, but with different cash/debt profits. In particular, the implications for cash-rich groups would be significant, in that if they have little or no external interest expense. Financing a subsidiary by a loan could result in a borrower not receiving a tax deduction for the interest even where the lender is in a higher tax jurisdiction than the borrower, and the loan had no tax benefit to the group. 4.6 The proposal would also create the situation where a company funded by a loan from a foreign bank could deduct all its interest, but the same company funded by a foreign parent might not be able to deduct any interest; a clear contravention of the arm s length principle. 4.7 Some of these effects could in part be mitigated by the proposed combined approach discussed in Part X of the discussion draft. However situations where a group is funding a large capital project, which is expected to be highly cash generative and economically can support a substantial amount of interest, and gives rise to a high debt/equity ratio in the company undertaking the project would not be covered by the combined approach. 4.8 As discussed above, we take the view that the BEPS effects of debt financing can be dealt with by CFC and transfer pricing rules, along with actions on matters such as hybrids. Work on harmful tax regimes to eliminate financing rulings with little economic rationale also has a part to play. 4.9 However, if, following this consultation exercise, the OECD nonetheless concludes that some form of interest restriction rule is required, a better approach would be a limit on interest deductibility across a group by reference to what the group could borrow, not what it has borrowed. This would prevent discrimination between highly externally leveraged groups and well capitalised groups. 4.10 Such a limit on interest deductibility should be subject to two levels of test: 4.10.1 at the level of the entity paying the interest, it should be shown that the interest payment meets an arm s length standard test; and 4.10.2 a fixed ratio approach a fixed ratio is adopted for all entities, subject to a P/tech/subsfinal/IT/2015 4

maximum limit across the group for deductions equating to the fixed ratio applied to the relevant group figure. This fixed ratio would be related to the Group s borrowing capacity, not by its actual borrowing. 4.11 A group having net interest expense in excess of the interest cap percentage should be allowed to carry forward the excess interest, as with other interest disallowances arising. 4.12 Examples are attached showing how this approach would affect groups with various levels of leverage. What is important is that they show that the outcomes are much more closely grouped using this combined approach than by allocating a cap based on actual interest expense, and correcting internal financing to avoid disallowances can be done whilst maintaining inter-company debt financing. Adopting this approach would mean that discrimination between groups is significantly reduced without creating base erosion. 5 Part IV. What is interest and what are payments economically equivalent to interest? 5.1 We are cautious about what payments/expenses might be included on the basis that they are economically equivalent to interest. In our experience such other things are often not all that comparable when other considerations are taken into account. In this context, we think it is very important that any test which seeks to include other payments/expenses involves both limbs in paragraph 34 of the discussion draft. That is the payment/expense is: (a) linked to financing AND (b) calculated by reference to an interest rate. 5.2 Ensuring there is a link to financing would (properly) exclude the following situations from the proposed rule: (i) a trader agrees to buy an asset say some land with completion due 1 January 2015. As the trader is slow getting its finance in order, it pays interest at the agreed daily rate on the overdue completion monies; (ii) the trader sells goods and offers a discount for advance payment calculated by reference to an interest rate; and (iii) damages are awarded against the trader subject to judgement interest. 5.3 It would also be useful to see an acknowledgement that a payment that is restricted by reference to the performance of the overall business ie a dividend on a share in a jurisdiction that has a distributable profits rule is never economically equivalent to interest. This is because it is a hallmark of interest that it is payable whether or not the business is profitable and whether or not the borrower is able to pay it. The two limbs test above does not seem to exclude fixed rate dividends. However, we suggest that fixed rate dividends should not be taken treated as interest: it would seem incorrect to treat non-deductible items as interest for the purposes of a deduction limitation rule. P/tech/subsfinal/IT/2015 5

5.4 This leads us to suggest that there should be a third limb to the test that is an amount is only economically equivalent to interest if it is tax deductible on no less favourable a basis in the relevant jurisdiction than interest would have been. If it does not command the same tax treatment, it isn t economically equivalent. 5.5 With regard to the payments identified in paragraph 35 of the discussion draft, we have concerns about the final three bullet points. 5.6 Foreign Exchange (FX) gains and losses many groups enter into hedging arrangements of various kinds to neutralise the impact of foreign exchange fluctuations. Many tax regimes have already developed detailed rules to accommodate this and prevent abuse. It is very difficult to see how these regimes could be aligned with the type of measures proposed in the consultation document. FX gains and losses should not be considered interest for these purposes. 5.7 Guarantee fees, arrangement fees and similar - whilst these amounts are often added to finance costs such as interest for the purposes of reporting, they are auxiliary and contractual amounts required for the finance to be provided, and are generally not seen as high risk items used for base erosion. To the extent that such items occur as inter-group payments, they are better dealt with from a transfer pricing perspective, and should not be dealt with in this action. 5.8 Question 1 refers to amounts incurred with respect to Islamic finance as an example of, presumably, economic equivalent payments. It would be helpful to clarify what other types of payments Working Party No 11 has in mind. 6 Part V. Who should a rule apply to? 6.1 We suggest that a 25% control test to determine whether entities are related is too high. 25% control is no control at all if one other shareholder has 75% control. Equally, if there are four 25% shareholders, one of them on their own has no power to force things through against the wishes of the others. We note that the 3 rd bullet point of paragraph 38 of the discussion draft actually refers to a 25% investment, and not control. Is this intended to pick up relationships involving loan stocks etc. as well as shares? If this is the case, then the result could be that the rules would capture things which should really fall within scenario 4 ; for example where a bank has loaned an amount which represents 25% of the total financing or conceivably where there is a Eurobond in issue which is held by a depositary for a clearing system or where a bond trustee holds rights on behalf of investors. 6.2 We suggest that using a 25% test and definition similar to the definition of related parties for the purposes of anti-hybrid rules recommended under Action 2, is not appropriate in these circumstances. The anti-hybrids rules are designed to negate benefits to people who would otherwise be getting them and structures to access hybrid benefits do not generally happen outside fully controlled groups. However an interest disallowance rule that works mechanically without enquiring whether the arrangements have in fact been engineered to produce a BEPS effect is rather different; the rule could have quite harsh and/or unexpected effects in commercial situations, particularly where capital/financing structures have been entered into without thinking about the tax consequences or in the context of joint ventures between unrelated parties. P/tech/subsfinal/IT/2015 6

6.3 We would therefore suggest a higher level of control to determine whether parties are related. 7 Part VI What should a rule apply to? 7.1 The conclusion in paragraph 45 of the discussion draft demonstrates the bluntness of the proposed interest limitation rule. Consider a company A, which is financially sound and is making good profits; the interest rate on which it can borrow funds is likely to be lower than company B which is struggling and thus pays more interest on the same amount of debt and has lower profits. If the restriction looks at debt by reference to EBITDA, company B is more likely to be denied a deduction than company A simply because it is struggling commercially and not because it is indulging in BEPS. We do not think that it can be right to have rules which restrict a company s deductions and, thereby increase its tax bill at a time when it is struggling. 7.2 Although it could be argued that the business should not be so highly geared, we are considering a situation where the debt was supportable at the time it was borrowed but a shock to the trade has changed things. There is a risk that a badly-designed BEPS Action 4 rule could apply with increasing harshness the more the company struggled. Paragraphs 51 and 52 of the discussion draft suggest that a company would not have to be large to be within the rule, so it could apply to companies with low profitability. 8 Part VII. Should a small entity exception or threshold apply? 8.1 We suggest that, at least for an initial period of, say, seven years the new rules should apply only to large groups (as with UK Debt cap) so that the cost of addressing any initial issues arising on the implementation of any new rules and the expertise to do so is provided by those groups that have most resources. Such an exception for small entities would also be helpful for tax authorities as it would mean that the number of groups that they have to deal with is smaller and they can concentrate their resources. 8.2 We note the point made in the first sentence of paragraph 56 of the discussion draft, that if a group is close to a country s threshold for moving into the new BEPS interest limitation, it may decide not to locate its new factory there. We agree that this is a concern and suggest that where, as in this case, a BEPS solution may end up distorting commercial decision making and capital ownership neutrality (see paragraph 3) then one has to query whether the approach is the correct one. Although the concern in paragraph 56 can be addressed by not having a threshold to entry to the new rule, the same concerns will arise if not all States agree to adopt the new rules, fearing a cost to jobs and inward investment etc. 9 Part VIII. Whether interest deductions should be limited with reference to the position of an entity s group 9.1 Part VIII of the discussion draft begins to consider the considerable accounting and compliance issues that would result from the proposals. At the very least any rule that allocates group debt, would need to have a mechanism that takes into account different types of businesses within a group. For example, if you had a group where P/tech/subsfinal/IT/2015 7

some companies owned land, other companies carried on farming trades and other companies engaged in commodity trading, there might have to be some acknowledgement of the different types of debt levels/interest costs incurred by the different types of trades in allocating group debt rather than allocating on a strictly pro rata basis. We would anticipate that this area would be one where considerable effort would be required to make an interest limitation work. 10 Part IX. Whether interest deductions should be limited with reference to a fixed ratio 10.1 We agree that there is very little to be said in favour of a fixed ratio rule on its own other than simplicity. It has the potential to produce the distortion referred to in paragraph 176 of the discussion draft. However, we do take the view that a ratio rule related to a Group s ability to borrow, rather than actual borrowing, would be a better approach for any interest restriction rule. 11 X. Whether a combined approach could be applied 11.1 Approach 1 is effectively using a fixed ratio test as a gateway to a more complex test. This seems quite a practical approach, within the context of a limitation of interest by reference to external borrowing (and thus does not deal with our fundamental points discussed above). However approach 2 seems to impose a higher compliance burden on groups with a poor credit rating (implying high interest and lower profits). These are just the sort of companies which do not need to incur extra compliance costs. 12 Part XI. The role of targeted rules 12.1 The discussion on targeted rules, comes back to the initial question, which is: what is BEPS Action 4 trying to achieve? Is it to prevent too much interest expense in the entity in question (as measured against some form of yardstick for the right amount) or is it trying to prevent interest expense arising in an entity for the wrong reasons? If the entity has no more interest expense than permitted by the yardstick, does it matter why it has the interest expense? If the reason for having the interest expense is important, then why are deductions for good interest (ie interest borne for bona fide commercial reasons) denied by the yardstick? 12.2 Please refer to the circumstances discussed at paragraph 7 above for an example of situations in which the rule as proposed could operate harshly. 13 Part XII. The treatment of non-deductible interest expense and double taxation 13.1 Under BEPS 2 (Hybrid mismatch) the OECD recommend rules to prevent asymmetries which lead to double deductions or deduction/non-recognition. It is disappointing that in BEPS 4 the OECD is not prepared to recommend rules that prevent double taxation. Indeed the application of interest cap as outlined would lead to double taxation in many cases. P/tech/subsfinal/IT/2015 8

13.2 There is no need to take the step of re-characterising the amount as a dividend if that leads to complications (as indicated in paragraphs 189-191 of the discussion draft), it would be sufficient to recommend that the rules in the other country do not impose a tax charge. This linked approach is of course the one adopted for Hybrid Mismatches. This is particularly important if BEPS Action 4 is going to produce a general rule that does not look at the purpose of the interest flow in question. 13.3 With regard to carry forwards, in principle it seems odd when the interest in question might be over the limit, but not produced by any form of tax avoidance (it just failed the yardstick test) to have an arbitrary cut-off. It may be that one group takes longer to turn its operations around than another or that one industry sector has different needs from another. 13.4 If the interest disallowance is a function of interest paid and EBITDA or some other profit metric, and there is no tax-avoidance criterion, interest carry-forward would be essential otherwise start-ups would be in a very odd position (lots of interest, no profits yet) and so would any other company which had a major problem or experienced a recession. The same problem could also be faced by groups which invested in long term projects. Putting a cap on the number of years would disadvantage long term investments, 13.5 A further point in relation to carry forwards: even if an MNE retains the ability to carry forward an interest deduction following the implementation of rules under BEPS 2, the ability to utilise the carry forward will still be subject to domestic rules which may, effectively render the benefit worthless. For example, the UK rules only allow nontrading interest deductions to be off set against future non-trading interest (and other loan relationship) income. For a holding company which has interest expense, but whose income is non-taxable dividends, the deferred tax asset of the carried forward interest expense will not be recognised under most GAAPs as its potential utilisation is too remote; hence, commercially, it has no value on the balance sheet. 14 Part XIII. Considerations for groups in specific sectors 14.1 The discussion draft recognises that banks and insurance companies present particular issues. Although noted above, for all MNEs the decision to use debt is nearly always commercially driven, it is important to remember that for the insurance and banking sectors, commercial (and where applicable regulatory) rather than tax considerations drive funding structure and the issue of debt and, often govern where it can be located. Capital is particularly expensive for insurers and banks, as regulators force insurers and banks to hold high quality capital in excess of expected liabilities, and limit the amount and type of debt that may be included in regulatory capital. There is therefore a natural tension between the minimum amount of capital required by regulators and ratings agencies, and the maximum amount of capital an insurance and banking group can use effectively. In light of this we welcome the recognition that specific rules should be developed for these sectors. 14.2 REITs in the UK are subject to a special tax regime which includes restrictions on interest allowed through the financing cost ratio rules (in Corporation Tax Act 2010 section 543). In addition the profits of the property rental business calculated under tax rules are not subject to corporation tax but instead there is a requirement that 90% of the profits are distributed to shareholders. P/tech/subsfinal/IT/2015 9

14.3 The majority of REITs have gearing between 25% and 40%. If a fixed ratio rule was introduced with a lower percentage no tax payment would result but the profits required to be distributed may increase. A REIT could therefore have insufficient reserves to make the distribution required, particularly if they were highly geared. We suggest that REITs are carved out of these provisions as they already have an interest cover test as described above. 15 The Chartered Institute of Taxation 15.1 The Chartered Institute of Taxation (CIOT) is the leading professional body in the United Kingdom concerned solely with taxation. The CIOT is an educational charity, promoting education and study of the administration and practice of taxation. One of our key aims is to work for a better, more efficient, tax system for all affected by it taxpayers, their advisers and the authorities. The CIOT s work covers all aspects of taxation, including direct and indirect taxes and duties. Through our Low Incomes Tax Reform Group (LITRG), the CIOT has a particular focus on improving the tax system, including tax credits and benefits, for the unrepresented taxpayer. The CIOT draws on our members experience in private practice, commerce and industry, government and academia to improve tax administration and propose and explain how tax policy objectives can most effectively be achieved. We also link to, and draw on, similar leading professional tax bodies in other countries. The CIOT s comments and recommendations on tax issues are made in line with our charitable objectives: we are politically neutral in our work. The CIOT s 17,000 members have the practising title of Chartered Tax Adviser and the designatory letters CTA, to represent the leading tax qualification. The Chartered Institute of Taxation 6 February 2015 P/tech/subsfinal/IT/2015 10

Example A Interest Cap Rule - Cap - 30% earnings "Normally leveraged" Group Bank Owes Bank - pays 25m interest A. Co Earnings 10m B owes A, pays 24m interest Revised interest payment 30m B. Co Earnings 45m C owes B, pays 18m interest Revised interest payment 16.5m Bank Deposit - earns 3m interest C. Co Earnings 45m BASE FIGURES Earnings 10.00 45.00 45.00 100.00 Entity Cap 3.00 13.50 13.50 Net third party interest income/(expense) (25.00) 0.00 3.00 (22.00) Allocation of overriding cap 3.00 13.50 13.50 30.00 Interest paid 25.00 24.00 18.00 Interest received 24.00 18.00 3.00 Net Interest income (expense) (1.00) (6.00) (15.00) (22.00) Entity Cap (3.00) (13.50) (13.50) (30.00) Interest allowed (1.00) (6.00) (13.50) (20.50) Net third party interest (22.00) Interest disallowed 1.50 ALTERNATIVE FIGURES Earnings 10.00 45.00 45.00 100.00 Entity Cap 3.00 13.50 13.50 Net third party interest expense (25.00) 0.00 3.00 (22.00) Allocation of overriding cap 3.00 13.50 13.50 30.00 Interest paid 25.00 30.00 16.50 Interest received 30.00 16.50 3.00 Net Interest income (expense) 5.00 (13.50) (13.50) (22.00) Entity Cap (3.00) (13.50) (13.50) (30.00) Interest allowed 0.00 (13.50) (13.50) (27.00) Interest taxed 5.00 5.00 Net interest allowed (22.00) Net third party interest (22.00) Interest disallowed 0.00

Example B Interest Cap Rule - Cap - 30% earnings Cash Rich Group A. Co Earnings 10m B owes A, pays 24m interest 30 B. Co Earnings 45m C owes B, pays 18m interest 16.5 Bank Deposit - earns 3m interest C. Co Earnings 45m BASE FIGURES Earnings 10.00 45.00 45.00 100.00 Entity Cap 3.00 13.50 13.50 Net third party interest expense 0.00 0.00 3.00 3.00 Allocation of overriding cap 3.00 13.50 13.50 30.00 Interest paid 0.00 24.00 18.00 Interest received 24.00 18.00 3.00 Net Interest income (expense) 24.00 (6.00) (15.00) 3.00 Entity Cap (3.00) (13.50) (13.50) (30.00) Interest allowed 0.00 (6.00) (13.50) (19.50) Interest taxed 24.00 24.00 Net interest taxed/(allowed) 4.50 Net third party interest 3.00 Interest disallowed 1.50 ALTERNATIVE FIGURES Earnings 10.00 45.00 45.00 100.00 Entity Cap 3.00 13.50 13.50 Net third party interest expense 0.00 0.00 3.00 3.00 Allocation of overriding cap 3.00 13.50 13.50 30.00 Interest paid 0.00 30.00 16.50 Interest received 30.00 16.50 3.00 Net Interest income (expense) 30.00 (13.50) (13.50) 3.00 Entity Cap (3.00) (13.50) (13.50) (30.00) Interest allowed 0.00 (13.50) (13.50) (27.00) Interest taxed 30.00 30.00 Net interest Taxed/(allowed) 3.00 Net third party interest 3.00 Interest disallowed 0.00

Example C Interest Cap Rule - Cap - 30% earnings Highly Leveraged Group Bank Owes Bank - pays 37m interest A. Co Earnings 10m B owes A, pays 24m interest 30 B. Co Earnings 45m C owes B, pays 18m interest 16.5 Bank Deposit - earns 3m interest C. Co Earnings 45m BASE FIGURES Earnings 10.00 45.00 45.00 100.00 Entity Cap 3.00 13.50 13.50 Net third party interest expense (37.00) 0.00 3.00 (34.00) Allocation of overriding cap 3.00 13.50 13.50 30.00 Interest paid 37.00 24.00 18.00 Interest received 24.00 18.00 3.00 Net Interest income (expense) (13.00) (6.00) (15.00) (34.00) Entity Cap (3.00) (13.50) (13.50) (30.00) Interest allowed (3.00) (6.00) (13.50) (22.50) Interest taxed 0.00 Net interest taxed/(allowed) (22.50) Net third party interest (34.00) Interest disallowed 11.50 ALTERNATIVE FIGURES Earnings 10.00 45.00 45.00 100.00 Entity Cap 3.00 13.50 13.50 Net third party interest expense 0.00 0.00 3.00 3.00 Allocation of overriding cap 0.00 0.00 0.00 0.00 Interest paid 37.00 30.00 16.50 Interest received 30.00 16.50 3.00 Net Interest income (expense) (7.00) (13.50) (13.50) (34.00) Entity Cap (3.00) (13.50) (13.50) (30.00) Interest allowed (3.00) (13.50) (13.50) (30.00) Interest taxed 0.00 Net interest Taxed/(allowed) (30.00) Net third party interest (34.00) Interest disallowed 4.00

Example D Interest Cap Rule - Cap - 30% earnings "Normally leveraged" Group, one company loss making Bank Owes Bank - pays 25m interest A. Co Earnings 10m B owes A, pays 42m interest 31 Deposit - earns 3m interest B. Co Earnings 125m Bank C. Co Earnings - 35m BASE FIGURES Earnings 10.00 125.00 (35.00) 100.00 Entity Cap 3.00 37.50 Net third party interest expense (25.00) 3.00 (22.00) Allocation of overriding cap 2.22 27.78 30.00 Interest paid 25.00 42.00 Interest received 42.00 3.00 Net Interest income (expense) 17.00 (39.00) 0.00 (22.00) Entity Cap (2.22) (27.78) 0.00 (30.00) Interest allowed 0.00 (27.78) 0.00 (27.78) Interest taxed 17.00 17.00 Net interest taxed/(allowed) (10.78) Net third party interest (22.00) Interest disallowed 11.22 ALTERNATIVE FIGURES Earnings 10.00 125.00 (35.00) 100.00 Entity Cap 3.00 37.50 (10.50) Net third party interest expense 0.00 0.00 3.00 3.00 Allocation of overriding cap 2.22 27.78 30.00 Interest paid 25.00 31.00 Interest received 31.00 3.00 Net Interest income (expense) 6.00 (28.00) 0.00 (22.00) Entity Cap (2.22) (27.78) (30.00) Interest allowed 0.00 (27.78) 0.00 (27.78) Interest taxed 6.00 6.00 Net interest Taxed/(allowed) (21.78) Net third party interest (22.00) Interest disallowed 0.22