HMRC has implied that the new interest restrictions from BEPS Action 4 should apply to banks. But no-one – including the OECD – knows how to make such ratios work in the context of their status as net interest profit makers and their regulatory constraints. Will we exclude them from the rules, apply the new rules anyway (to the detriment of banks in trouble), or just target specific scenarios? What should we do about non-banking members of banking groups? And, absent of clear guidance from the OECD, how can the new regime come into force from 1 April 2017?
On 28 July 2016, the OECD published a paper with the catchy title BEPS Action 4: Approaches to address BEPS involving interest in the banking and insurance sectors. It was supposed to be the long-awaited answer to the question of how to tackle the (alleged) problem of base erosion and profit shifting (BEPS) involving interest deductibility in those sectors, to give G20 fiscal authorities (including HMRC) a guide to what they needed to do.
Unfortunately, anyone expecting real enlightenment from the OECD on this point will have been sadly disappointed. If one felt uncharitable, one could sum up the 33 pages in this single phrase: ‘We don’t know, perhaps you could tell us?’
In fairness to the OECD, it sets itself an impossible question, so it is no surprise that they are having trouble answering it. To understand why they are having problems, the few proposals they do make, and what it means for banks in the UK, a few basics need to be borne in mind.
Once upon a time (a couple of years ago), the OECD published an action plan, proposing 15 actions designed to combat BEPS at an international level. Action 4 on interest deductibility focuses on interest deductions as a bad thing, and proposes to limit BEPS by a general rule introduced to restrict the deductibility of interest payments, regardless of the purpose of the debt or the identity of the counter-party. The OECD’s stance on the matter is that: ‘The mobility and fungibility of money makes it possible for multinational groups to achieve favourable tax results by adjusting the amount of debt in a group entity.’ The fear is that international groups can manipulate their taxable income by ensuring that they (i) stuff lots of third party debt into high tax country operations; (ii) use intra-group loans to get higher interest deductions country by country than their third party debt levels actually generate; and (iii) use third party or intra-group finance to fund tax exempt income.
Action 4 is proving to be one of the most significant actions for UK companies, because our current regime – once you get passed our plethora of anti avoidance – is generous to taxpayers. Generally, interest paid on debt financing is deductible from a company’s UK corporation taxable profits and therefore a company’s liability to UK corporation tax is reduced. This form of tax relief is often invaluable, but the concern is that it leads to scope for abuse. The example usually given is that the UK rules ‘let’ you load up companies in the UK – a higher tax jurisdiction – with high levels of debt to reduce taxable profits, whilst shifting business profits to low tax jurisdictions such as tax havens so little or no tax is paid (the fact that the UK’s existing anti-avoidance rules do not actually let you do that is sadly seen as irrelevant).
In response to this we have thus far seen from HMT and HMRC a high-level consultation agreeing with the OECD’s approach to introducing a fixed ratio interest limitation rule, an announcement in Budget 2016 that a rule limiting interest deductions would be introduced with effect from April 2017, and another consultation in May 2016 providing details of the proposed new regime. Further announcements are expected in Autumn Statement 2016, with draft legislation for inclusion in Finance Bill 2017 to follow shortly thereafter. Crucially, HMRC has already implied it wants to apply some sort of new rules to banks from 1 April 2017 alongside everyone else – without knowing what the OECD’s final recommendations for the sector will be.
So what do the new rules do?
There’s lots of detail on what the general rules will look like, but the two key concepts are the FRR and the GRR. It’s important to see how they work before considering how and why they don’t work in the banking sector.
Under the FRR, tax relief for interest and other financing costs is limited to 30% of EBITDA. EBITDA for these purposes is not the figure in the accounts; it is a special calculation of ‘tax-EBITDA’. Groups will need to work out the tax-EBITDA of each UK resident member company and each UK permanent establishment (PE), and add them together. 30% of that figure will be the limit on deductible interest.
Once you’ve worked that out, you then need to calculate the net interest expense of each UK group company and PE. Again, interest is not the accounting figure – it will be what is called ‘tax-interest’, which includes other finance costs and items economically equivalent to interest. Add up the total and if the group’s net tax-interest expenses exceed the 30% limit there will be an interest restriction equal to the excess.
There are other aspects to the rule: a group de minimis allowance of £2m per annum, the ability to carry forward indefinitely interest expenses restricted under the rule, and carry forward for three years any spare interest capacity, but no carry back and no grandfathering. And just in case you were thinking this will be easy, a modified version of the existing ‘worldwide debt cap’ rules will also sit alongside everything else.
If you get trapped by the FTT, you can use the GRR to get out of the rules as an alternative measure. It is currently proposed that the group ratio will be calculated by dividing the net qualifying group interest expense by group-EBITDA. In applying the GRR, net interest is calculated in a similar way as for the FRR. The worldwide ‘group-EBITDA’ is broadly the accounting measure, although that too is adjusted for certain items. The amount of deductions available under the GRR will be capped at the amount of net qualifying group-interest expense.
Now for the tricky bit. The FRR and GRR simply do not work for banks. As the OECD themselves admit, since banks are entities for whom interest is the breath of life, they tend to make net interest profits, so when you try to apply the formulae you don’t get any further than calculating the net interest expense – simply because there isn’t any. As the OECD also note, the only real life circumstance where that reverses is when market conditions are poor, so a bank finds itself with net interest expenses; but applying the FRR and disallowing most or all of this net expense would seriously hinder a bank’s ability to survive financial shocks – the very moment you don’t want to hinder them and cause another financial crisis.
So, objection number one to applying the FRR and GRR to banks is that it simply won’t work for them, so why make them jump through the hoops each year for the sake of it?
Objection number two is one of policy: how likely are banks to fall foul of BEPS Action 4? Banks make their money from the turn between borrowing and lending. That means they are subject to a host of regulatory restrictions about what they can and cannot do, so (a) it is pretty hard for them to get over-leveraged (although the OECD do try to make the case for instances where that can happen, especially in branches); and (b) restructuring their operations if they accidentally fall foul of these rules is difficult. And they are also of course one of Britain’s biggest businesses. With that in mind, restricting their interest relief across the board would be a particularly odd move even if the FRR and GRR did work in relation to them.
The problem HMRC has is in not committing to exclude banks from the rule, so what form should it take for them? They may have expected the OECD to come back with some cogent proposals to tell them what to do – if so, the OECD has let them down. The evident lack of progress in the July paper is disappointing, in that it basically takes 33 pages for the OECD to conclude that they don’t know what to conclude.
The OECD do not think that banks (or insurance companies) pose a high risk in relation to BEPS involving interest and therefore do not think that there is a need to develop a single approach that countries should adopt across the board. Unsurprisingly, they also agree that the FRR and the GRR are probably not appropriate tools to combat any risks which governments may identify. And they have rather parked the problem by saying their final recommendations (which are expected by the end of 2016) will look at the approaches applied by different countries, presumably in the hope that someone else will solve the mess and come up with a way forward for banks, so they don’t have to.
Given this, the preferred option from the OECD’s various suggestions is to exclude banks and insurance companies from the rules altogether. We live in hope, but it’s a slim chance HMRC will go for that option.
In relation to the FRR, the OECD paper details five examples of how groups containing banks could be treated. One option considered is to exclude banks when the fixed ratio rule is applied to a local group, so that net interest income in banks cannot effectively allow a higher interest deduction by non-banking entities in the group. Another option discussed is whether to carve out special purpose vehicles that issue debt into the market to support banking activities. Targeted anti-avoidance rules (TAARS) for specific scenarios are also advocated in a big way.
In relation to the GRR, the OECD outlines another several options for dealing with banks which countries can consider, including the disapplication of the GRR altogether or allowing the GRR for banking groups but excluding from the calculation items related to the funding and results of the banks in the group.
The paper is a mish-mash of conflicting ideas, but two points come through: (i) it is all very complicated; and (ii) the OECD doesn’t know what to do, so it is for G20 fiscal authorities to decide what they want to do, and then share that with the OECD. It’s a bit like pick and mix, only the matter at stake is the global financial economy instead of sweets.
We wait to see what HMRC’s response will be in the Autumn Statement. At the very least I gloomily predict we can expect yet more TAARs aimed at particular issues the OECD pinpoint, such as payments of interest on ‘artificial loans’ where no new funding is raised by the entity or its group, and payments of interest to related parties that is used to finance the production of tax-exempt income. We may also see some rules dealing with applying the FRR and GRR to companies in banking groups who are not themselves banks (i.e. excluding the bank but catching everyone else in the group). Though how you work out what an artificial loan is, or which bit of fungible loan finances which bit of income, or what you do about investment banking and other financial activities in how you define a ‘bank’, will be interesting to see. It would be good if it stays away from certain other OECD identified risks, such as payments of excessive interest and payments to related parties who are subject to low taxation on the interest income – which are already hammered by transfer pricing and our other existing anti-avoidance rules, before we suffocate under the weight of extra TAARs.
Whatever HMRC has planned, here’s hoping the department is already on top of it, because it does not have long if it really does want to stick to that 1 April 2017 commencement date.
HMRC has implied that the new interest restrictions from BEPS Action 4 should apply to banks. But no-one – including the OECD – knows how to make such ratios work in the context of their status as net interest profit makers and their regulatory constraints. Will we exclude them from the rules, apply the new rules anyway (to the detriment of banks in trouble), or just target specific scenarios? What should we do about non-banking members of banking groups? And, absent of clear guidance from the OECD, how can the new regime come into force from 1 April 2017?
On 28 July 2016, the OECD published a paper with the catchy title BEPS Action 4: Approaches to address BEPS involving interest in the banking and insurance sectors. It was supposed to be the long-awaited answer to the question of how to tackle the (alleged) problem of base erosion and profit shifting (BEPS) involving interest deductibility in those sectors, to give G20 fiscal authorities (including HMRC) a guide to what they needed to do.
Unfortunately, anyone expecting real enlightenment from the OECD on this point will have been sadly disappointed. If one felt uncharitable, one could sum up the 33 pages in this single phrase: ‘We don’t know, perhaps you could tell us?’
In fairness to the OECD, it sets itself an impossible question, so it is no surprise that they are having trouble answering it. To understand why they are having problems, the few proposals they do make, and what it means for banks in the UK, a few basics need to be borne in mind.
Once upon a time (a couple of years ago), the OECD published an action plan, proposing 15 actions designed to combat BEPS at an international level. Action 4 on interest deductibility focuses on interest deductions as a bad thing, and proposes to limit BEPS by a general rule introduced to restrict the deductibility of interest payments, regardless of the purpose of the debt or the identity of the counter-party. The OECD’s stance on the matter is that: ‘The mobility and fungibility of money makes it possible for multinational groups to achieve favourable tax results by adjusting the amount of debt in a group entity.’ The fear is that international groups can manipulate their taxable income by ensuring that they (i) stuff lots of third party debt into high tax country operations; (ii) use intra-group loans to get higher interest deductions country by country than their third party debt levels actually generate; and (iii) use third party or intra-group finance to fund tax exempt income.
Action 4 is proving to be one of the most significant actions for UK companies, because our current regime – once you get passed our plethora of anti avoidance – is generous to taxpayers. Generally, interest paid on debt financing is deductible from a company’s UK corporation taxable profits and therefore a company’s liability to UK corporation tax is reduced. This form of tax relief is often invaluable, but the concern is that it leads to scope for abuse. The example usually given is that the UK rules ‘let’ you load up companies in the UK – a higher tax jurisdiction – with high levels of debt to reduce taxable profits, whilst shifting business profits to low tax jurisdictions such as tax havens so little or no tax is paid (the fact that the UK’s existing anti-avoidance rules do not actually let you do that is sadly seen as irrelevant).
In response to this we have thus far seen from HMT and HMRC a high-level consultation agreeing with the OECD’s approach to introducing a fixed ratio interest limitation rule, an announcement in Budget 2016 that a rule limiting interest deductions would be introduced with effect from April 2017, and another consultation in May 2016 providing details of the proposed new regime. Further announcements are expected in Autumn Statement 2016, with draft legislation for inclusion in Finance Bill 2017 to follow shortly thereafter. Crucially, HMRC has already implied it wants to apply some sort of new rules to banks from 1 April 2017 alongside everyone else – without knowing what the OECD’s final recommendations for the sector will be.
So what do the new rules do?
There’s lots of detail on what the general rules will look like, but the two key concepts are the FRR and the GRR. It’s important to see how they work before considering how and why they don’t work in the banking sector.
Under the FRR, tax relief for interest and other financing costs is limited to 30% of EBITDA. EBITDA for these purposes is not the figure in the accounts; it is a special calculation of ‘tax-EBITDA’. Groups will need to work out the tax-EBITDA of each UK resident member company and each UK permanent establishment (PE), and add them together. 30% of that figure will be the limit on deductible interest.
Once you’ve worked that out, you then need to calculate the net interest expense of each UK group company and PE. Again, interest is not the accounting figure – it will be what is called ‘tax-interest’, which includes other finance costs and items economically equivalent to interest. Add up the total and if the group’s net tax-interest expenses exceed the 30% limit there will be an interest restriction equal to the excess.
There are other aspects to the rule: a group de minimis allowance of £2m per annum, the ability to carry forward indefinitely interest expenses restricted under the rule, and carry forward for three years any spare interest capacity, but no carry back and no grandfathering. And just in case you were thinking this will be easy, a modified version of the existing ‘worldwide debt cap’ rules will also sit alongside everything else.
If you get trapped by the FTT, you can use the GRR to get out of the rules as an alternative measure. It is currently proposed that the group ratio will be calculated by dividing the net qualifying group interest expense by group-EBITDA. In applying the GRR, net interest is calculated in a similar way as for the FRR. The worldwide ‘group-EBITDA’ is broadly the accounting measure, although that too is adjusted for certain items. The amount of deductions available under the GRR will be capped at the amount of net qualifying group-interest expense.
Now for the tricky bit. The FRR and GRR simply do not work for banks. As the OECD themselves admit, since banks are entities for whom interest is the breath of life, they tend to make net interest profits, so when you try to apply the formulae you don’t get any further than calculating the net interest expense – simply because there isn’t any. As the OECD also note, the only real life circumstance where that reverses is when market conditions are poor, so a bank finds itself with net interest expenses; but applying the FRR and disallowing most or all of this net expense would seriously hinder a bank’s ability to survive financial shocks – the very moment you don’t want to hinder them and cause another financial crisis.
So, objection number one to applying the FRR and GRR to banks is that it simply won’t work for them, so why make them jump through the hoops each year for the sake of it?
Objection number two is one of policy: how likely are banks to fall foul of BEPS Action 4? Banks make their money from the turn between borrowing and lending. That means they are subject to a host of regulatory restrictions about what they can and cannot do, so (a) it is pretty hard for them to get over-leveraged (although the OECD do try to make the case for instances where that can happen, especially in branches); and (b) restructuring their operations if they accidentally fall foul of these rules is difficult. And they are also of course one of Britain’s biggest businesses. With that in mind, restricting their interest relief across the board would be a particularly odd move even if the FRR and GRR did work in relation to them.
The problem HMRC has is in not committing to exclude banks from the rule, so what form should it take for them? They may have expected the OECD to come back with some cogent proposals to tell them what to do – if so, the OECD has let them down. The evident lack of progress in the July paper is disappointing, in that it basically takes 33 pages for the OECD to conclude that they don’t know what to conclude.
The OECD do not think that banks (or insurance companies) pose a high risk in relation to BEPS involving interest and therefore do not think that there is a need to develop a single approach that countries should adopt across the board. Unsurprisingly, they also agree that the FRR and the GRR are probably not appropriate tools to combat any risks which governments may identify. And they have rather parked the problem by saying their final recommendations (which are expected by the end of 2016) will look at the approaches applied by different countries, presumably in the hope that someone else will solve the mess and come up with a way forward for banks, so they don’t have to.
Given this, the preferred option from the OECD’s various suggestions is to exclude banks and insurance companies from the rules altogether. We live in hope, but it’s a slim chance HMRC will go for that option.
In relation to the FRR, the OECD paper details five examples of how groups containing banks could be treated. One option considered is to exclude banks when the fixed ratio rule is applied to a local group, so that net interest income in banks cannot effectively allow a higher interest deduction by non-banking entities in the group. Another option discussed is whether to carve out special purpose vehicles that issue debt into the market to support banking activities. Targeted anti-avoidance rules (TAARS) for specific scenarios are also advocated in a big way.
In relation to the GRR, the OECD outlines another several options for dealing with banks which countries can consider, including the disapplication of the GRR altogether or allowing the GRR for banking groups but excluding from the calculation items related to the funding and results of the banks in the group.
The paper is a mish-mash of conflicting ideas, but two points come through: (i) it is all very complicated; and (ii) the OECD doesn’t know what to do, so it is for G20 fiscal authorities to decide what they want to do, and then share that with the OECD. It’s a bit like pick and mix, only the matter at stake is the global financial economy instead of sweets.
We wait to see what HMRC’s response will be in the Autumn Statement. At the very least I gloomily predict we can expect yet more TAARs aimed at particular issues the OECD pinpoint, such as payments of interest on ‘artificial loans’ where no new funding is raised by the entity or its group, and payments of interest to related parties that is used to finance the production of tax-exempt income. We may also see some rules dealing with applying the FRR and GRR to companies in banking groups who are not themselves banks (i.e. excluding the bank but catching everyone else in the group). Though how you work out what an artificial loan is, or which bit of fungible loan finances which bit of income, or what you do about investment banking and other financial activities in how you define a ‘bank’, will be interesting to see. It would be good if it stays away from certain other OECD identified risks, such as payments of excessive interest and payments to related parties who are subject to low taxation on the interest income – which are already hammered by transfer pricing and our other existing anti-avoidance rules, before we suffocate under the weight of extra TAARs.
Whatever HMRC has planned, here’s hoping the department is already on top of it, because it does not have long if it really does want to stick to that 1 April 2017 commencement date.
