Enter Your E-mail

Tolley's Practical Tax by Andrew Hubbard

Andrew Hubbard takes stock of the current situation regarding the disclosure rules

Introduction

It is a year since the announcement, in Budget 2004, of the introduction of the regime for early disclosure of tax avoidance schemes. The regime became live, after much controversy, in September 2004 and it is now time, after several months’ of practical experience, to take another look at how effective the regime actually is and how it is operating in practice. I have already used the term ‘scheme’ once and I will continue to use it throughout this article, because it is a useful shorthand term. It must be understood, however, that the regulations do not only cover packaged tax avoidance products: one-off bespoke planning can, and often will, fall within the scope of what is disclosable.

It might be useful first to give readers a brief reminder of the way that the regime operates. To begin with, it is important to appreciate that there are quite separate regimes for direct and indirect taxes: there are some points of similarity between these, but essentially, they are wholly independent of each other. This article is only concerned with the disclosure rules for direct tax.

The regime in outline

The essential feature of the regime is that it imposes the primary reporting obligation on the adviser rather than the user. An adviser who has devised an avoidance scheme is under an obligation to provide details of the scheme to the Revenue within five days of the scheme first being ‘made available for implementation’. There is a penalty of up to £5,000 for failure to make the necessary disclosure. The notification process is not an approval process: the Revenue will register the scheme regardless of whether or not it believes that it is effective. A promoter who sells a registered scheme to a client must provide that client with the reference number of the scheme, and the client must in turn include the reference number on his tax return. Where there is no UK-resident promoter of the scheme, or where the scheme is devised in house, the primary reporting obligation falls on the client. Where a lawyer asserts legal and professional privilege the primary obligation to disclose the scheme falls on the client.

It is vital to be clear that it is the scheme not the client which is disclosable. The Revenue wants to know about ways in which advisers have managed to exploit what the government perceives as loopholes in the legislation, so that in appropriate cases it can recommend to ministers that the legislation is changed. Once a scheme has been notified the Revenue has the information that it needs, and there is therefore no need for an adviser to notify the Revenue each time the same scheme is used. That process is covered by the requirement for the client to include details of the scheme on his or her tax return. Clearly there will be some marginal cases where what is done for one client is not precisely the same as that which is done for another client: the Revenue guidance (available on the AIU section of the website) has some useful material on how to decide when schemes are sufficiently different to require separate notification.

The scope of this article

I can now turn to some of the practical issues which have emerged in the first few months of the regime’s life. I do not want to say much here about mass marketed avoidance schemes. Practitioners who are involved in designing or promoting such schemes are generally in no doubt that what they are doing falls squarely within the reporting requirements, and there seems to be a general acceptance that the regime sets out the obligations in respect of such schemes fairly well.

I want to address myself instead to the position of what I imagine the typical reader of this article to be. That is somebody who is not designing and marketing tax avoidance products, but who has a strong advisory practice and is always looking to work with his or her clients to minimise taxation liabilities. Most of the time the practitioner will be doing what he or she regards as mainstream planning, but there will be times where he or she is doing something more adventurous and will be concerned that there is a danger of drifting into territory covered by the disclosure rules. Such advisers may well be working for a firm which does not have a large national technical department experienced in the nuances of the disclosure regime, and they may have to make judgements with very little to guide them.

Is there a disclosable scheme?

The most difficult issue for such practices will be deciding whether or not the advice which is being given falls into one of the specified categories of scheme which have to be disclosed. There are other problems of course, such as determining precisely when the disclosure is due, but it is this basic issue of actually knowing whether or not you are in the regime in the first place which is by far the most problematic.

The scope of the regulations is actually quite narrow. Although the primary legislation deals with virtually all of the direct taxes the regulations narrow these down to just three taxes: income tax, capital gains tax and corporation tax initially, plus stamp duty land tax from 1 July 2005. So there is no need (at least for the present) to be concerned about inheritance tax. Secondly, it is only two areas within the first three taxes which are within the regime.

Financial and employment products

The basic proposition is that it is only arrangements involving the use of ‘financial products’ and ‘employment products’ which are disclosable. So clearly we need to understand what these might be. And this is where all of the problems start. There is a clear policy objective to seek disclosure only of ‘abusive’ avoidance schemes, but of course there is no way of defining these in an objective way which will be accepted by everybody. So we are into classic ‘elephant’ territory. An elephant is something which we all recognise but would find it almost impossible to define. We have the same issue here.

The period between the original announcement of the regime and the day on which it became live saw a great deal of work being put in, on both sides, to come up with a workable definition. The Revenue was concerned that the definition had to be robust enough to ensure that avoidance schemes could not slip through the net on a technicality: the profession was equally concerned that under the regulations, as originally drafted, a huge amount of ‘ordinary’ planning would have to be disclosed.

Eventually a position was reached which nobody accepts is perfect but which at least can be operated without significant risk to the Revenue and without most straightforward planning being needlessly sucked into the net.

So after all of this activity what are we actually left with?

I mentioned earlier that the two key areas are employment products and financial products. We need to look at each of these in turn.

Employment products

There are three strands to employment products: securities or associated rights, payments to trustees or intermediaries, and loans. What the regulations are looking for are situations where one or more of these products is used in such a way that ‘a tax advantage by way of a reduction in, or deferment of, liability might be expected to be obtained, by virtue of the arrangements, by the employer or the employee, or by any other person, by reason of the employee’s employment.’

There are a number of important issues here. First of all there is no need for there actually to be a tax advantage. This might at first glance seem odd, but in fact it is logical. If I devise a planning scheme I hope that it will work, indeed I expect that it will, but if it is challenged it could take years for the Courts to rule on its effectiveness. Thus at the date that I promote the scheme I cannot say that it will give rise to a tax advantage. I could therefore legitimately not disclose the scheme, because strictly a requirement that the scheme will give rise to an advantage would not be met. So the test is the lesser one of ‘might be expected’. Since I will hardly promote a scheme which I do not expect to be successful I now have no argument against disclosure.

The second issue is that the tax advantage can arise to either the employer or employee. This is important. In many remuneration planning arrangements it will be the employer who implements the planning scheme but it will be employee who actually receives the tax advantage. The Revenue was concerned that, without this wide test, remuneration schemes would escape disclosure, because the person who implemented the scheme did not obtain a direct advantage from it. Similarly, the reference to advantages received by ‘any other person by reason of the employee’s employment’ is important. There have been a number of schemes where the employee benefits through indirect arrangements such as family trusts or benefits to ‘friendly’ third parties, and clearly the Revenue wanted to ensure that these were disclosable.

The three strands

We now need to look at what I called earlier the three strands of the employment products definition. The first, and probably the most important, is ‘securities or associated rights’. There is an exhaustive definition of these, which you will need to look at in cases of difficulty, but, for present purposes, you can assume that it will include all shares, securities and options over these, regardless of whether the securities are in the employing company or in an unrelated business. Much of the avoidance of tax on remuneration has been through the use of ever-more exotic forms of shares in special purpose companies, and this definition has been largely successful in ensuring that all such ‘exotica’ is disclosable. But don’t assume that it is only ‘funny’ shares which are caught. Straightforward ordinary shares in the employing company will be employment products.

The second and third strands can be dealt with more concisely. Payment to trustees and intermediaries is clearly primarily designed to catch employee trust arrangements and loans are largely self-evident. Some care does need to be taken with this one. Normal employment-related loans which come within the beneficial loan rules are excluded from disclosure. But some planning schemes have involved loans which, for all practical purposes, are intended to remain in place indefinitely. Where it is the intention of the parties that the ‘loan’ will not be repaid it is not exempt from disclosure.

The filters

Having created a very wide definition of employment products the regulations then narrow down the scope of what is reportable, through what are generally referred to as a series of filters, until all that is left are those things which the Revenue really wants to know about.

First of all there are filters within each definition. For example within the securities definition there is an exception for approved share incentives. This is common sense. A company implementing a SAYE scheme will be using an employment product, (ie the share option) but it would be absurd for this to be disclosable, given that all that the employer is doing is using an approved government incentive. EMI schemes themselves are not disclosable, but arrangements put in place to enable an EMI scheme to operate can be disclosable. For example a company paying a dividend to get its gross assets down to below the £30m limit will be within the disclosure regime. This is widely seen as an absurdity, but at the moment the legislation requires disclosure, unless one of the other filters removes the reporting obligation. There is also an explicit carve out for shares which qualify for EIS relief. Again this is simply common sense.

I have already mentioned the exception for loans which are taxed under the beneficial loan regime. For trusts and intermediaries the main carve out is for payments into approved pension schemes: this is again a sensible measure, which prevents unnecessary compliance burdens.

Although the filters within each specified product do start to narrow down the scope of what is disclosable, they still leave a huge amount which is still potentially within the regime.

Excluded arrangements

This is where the second, and most important, level of filter comes in. An employment product is not disclsosable if it is an ‘excluded arrangement’. This is an arrangement which involves neither a ‘premium’ fee or a ‘confidentiality’ arrangement. It is vital to understand these tests, particularly the former, because they are the key to understanding how the regime actually operates in practice.

A premium fee is a ‘a fee which is chargeable by virtue of any element of the arrangements (including the way in which they are structured) from which the tax advantage expected to be obtained arises, and is a fee the amount of which is, to a significant extent, attributable to that tax advantage or which is, to any extent, contingent upon the obtaining of that advantage’.

We need to dissect this complex definition with some care. The first point is that the ‘amount’ of the fee must be attributable to the tax advantage. Every time that an adviser issues a bill for tax advice the fee is, by definition, attributable to a tax advantage (you are hardly going to issue a fee for increasing your client’s tax liability!), but this is not what is meant here. The test is looking at situations where the size of the fee is influenced by the tax advantage. This is most obvious in cases where there is a direct linkage between the fee and the tax saving: ‘this scheme will save you £100K of tax – my fee will be 20% of the saving’. But it is clear that the definition is wider than this. Effectively what the Revenue is looking for are situations where the adviser is not billing by reference to hours worked but is instead seeking a fee for selling an idea or piece of intellectual property to a client.

There are still some considerable difficulties with all of this and we are expecting further guidance from the Revenue on this area shortly. In normal practice-management terminology we talk about getting a ‘premium fee’ where we achieve more than 100% of our normal charge out rate, but this is clearly not what is meant by a premium fee in the definition. There does, however, come a point where the fee ceases to be related to the amount of input and instead starts to be related to the effect that the planning has on the client’s overall tax position. That is when it becomes a premium fee. The current Revenue guidance has some useful examples of where the boundaries might lie, but these are not without their problems and the position is still not wholly satisfactory

Contingent fees

The next point to note is that the definition includes contingent fees. There are a couple of points here which need to be stressed. First of all the contingent fee has to be related to the tax advantage. This needs some explanation. It is common for corporate finance transactions such as management buyouts to be structured with contingent fees: if the transaction does not proceed the adviser does not get paid. This is not a contingent fee for these purposes. The contingency in an MBO is whether or not the transaction happens, not the obtaining of a particular tax result. Similarly, a capital allowances review, where the adviser gets a fee which depends on the amount of expenditure which can be claimed as plant and machinery, is not within the rules. The contingent element does not depend on the transaction – it depends on the adviser’s ability to understand the definition of plant. But if I devise some planning and I say to the client ‘if this succeeds I will bill you an extra £5,000’ I do have a contingent fee within the definition. It is important to note that ‘size’ and ‘contingency’ are separate. A structure in which an adviser would get a ‘normal’ fee if a contingency was satisfied would be within the regulations.

Reasonable expectation

It would be easy to get round the premium fee test by manipulating charge-out rates. If I expected that the client would save £100,000 and I wanted a 20% fee, I could simply work backwards and say that I would charge out my 10 hours work at £2,000 per hour, instead of my normal rate of, say, £200 per hour. The regulations prevent this by imposing a reasonable expectation test. In other words, it is only if there is a reasonable expectation that no promoter could obtain a premium fee from the client in similar circumstances that the advice is not disclosable. But it is also worth noting that the test assumes that the client is ‘experienced in receiving services of the type being promoted’. In other words, the fact that a particular promoter might be able to extract a premium fee for straightforward advice by, in effect, ripping off a little old lady would not mean that that advice would be considered as commanding a premium fee.

Does the premium fee test work?

Intellectually one can make many criticisms of the premium fee test. Indeed, at a wholly theoretical level one could never satisfy the premium fee test, because it seems to demand knowledge of how every tax idea would be priced by every adviser in the UK. At a pragmatic level, however, the test is probably as effective as anything else which could have been devised. The whole thrust of the disclosure regime is aimed at the adviser rather than the client and therefore it makes sense to use a supply-side test to determine what is or is not a premium fee. After all the promoter of a scheme should know better than anybody whether or not he has devised something novel and innovative for which a client will be prepared to pay a non-standard fee. That is essentially what the premium fee test is trying to do.

Confidentiality

There is one other element to the excluded arrangements test, which is entirely freestanding. If the tax advantage arises through planning which a ‘promoter might reasonably be expected to keep confidential from other promoters’ it will be disclosable.

Tax advisers are under a duty of confidentiality to their clients, but this is not what is envisaged here. The regulations are dealing with the situation where the promoter imposes a confidentiality condition on the client – ie prohibits the client from giving any other adviser details of the scheme. Again, this is entirely consistent with the thrust of the disclosure regime, which is directed at the adviser who wants to make maximum capital out of a scheme by ensuring that nobody else can market that scheme in competition.

Taken together, the various elements of the premium fee and excluded arrangements rules will remove the obligation for advisers to report many, if not most, straightforward tax planning arrangements. At the margins there are still problems with the precise way that the test operates, but it is a major advance on what was there before. Anecdotal evidence suggests that the definition is broadly achieving what it is intended to do.

The white list

Before the premium fee test was introduced the CIOT and the Revenue issued a ‘white list’ of planning which was agreed to be outside the reporting regime. To some extent this list has been superseded by events, and its precise status at the moment is the subject of some uncertainty, but it is still useful in getting an insight into the Revenue’s thinking. The full text of the white list is available on the CIOT website (www.tax.org.uk): in summary it covers flexible benefit arrangements, salary sacrifices, incorporations, dividends to employee shareholders, standard dual contract arrangements and bonus deferrals (the last two are mainly aimed at expatriate arrangements). Until revised guidance is issued the white list remains a useful adjunct to the Revenue’s official guidance.

Financial products

So far we have concentrated on employment products. We also need to deal with financial products. In the original regulations the test of what was disclosable was driven by a formula, which had to be applied separately to each financial product. This was widely seen as imposing an unreasonable burden on advisers and it also produced some very unusual outcomes: some highly aggressive planning schemes, well known in the market, would have slipped through the net because of the way that the formula differentiated between what was and was not disclosable. Eventually the formula-based approach was abandoned in favour of a premium fee approach.

Before we look at this we need to understand which products are included in the basic definition. The regulations specify a number of disclosable products. The most important of these are undoubtedly shares and loans. There are also more complex products, such as stock lending arrangements and repo agreements, but it is shares and loans which most advisers are likely to come across. There are detailed definitions for each product and in practice these will need to be examined, but it can generally be assumed that these are widely drawn. The definition of shares, for example, includes straightforward ordinary shares as well as more complex share arrangements.

As with employment products, there are some limited exclusions for tax favoured products: for example, PEPs and ISAs are not treated as financial products for disclosure purposes.

The filters

The premium fee and confidentiality filters apply to financial products in broadly the same way as they do for employment products. There are some differences at the margin, mainly to deal with the position of banks who price tax efficient financial products through a bid offer spread rather than via a fee, but as far as most advisers will be concerned it is reasonable to treat the premium fee and confidentiality tests as being the same for both sets of products.

Making the report

Once the decision has been made that something is reportable the next decision is how and when to make the report. The reporting deadline is five days. This is widely seen as impractical by most professionals but the time limit is a point of principle on which the government was unwilling to compromise. Effectively it means that the adviser needs to prepare the disclosure at the same time as giving the advice. The five-day period is triggered by the earlier of the scheme’s being made available for implementation and the actual implementation. There have been some problems over precisely what these tests mean, and the Revenue guidance does give some explanation about how it believes that this time limit operates. In cases of difficulty the guidance should be consulted, but in practice the adviser should assume that as soon as a client is in a position to agree to a particular piece of planning being implemented the notification clock will start to tick.

Completing the forms

The report itself is to be made on forms specified by the Revenue: these can be downloaded from the Revenue website. There is still very little guidance on exactly how much detail is required on the forms and the Revenue’s avoidance intelligence unit, which has responsibility for policing disclosures under the scheme, has had discussions with many firms to agree an appropriate level of disclosure. It is clear that the Revenue is not looking for huge amounts of detail or an exhaustive technical analysis of every single aspect of the planning. In essence it is seeking a level of disclosure which will make clear the particular ‘trick’ which allows the tax advantage to be obtained. In practice of course things are not quite that simple, and the author would recommend that any adviser who is unsure of quite how to make the disclosures to discuss his or her intended approach with the Revenue.

Conclusions and operating the scheme in practice

In the end we have now reached a situation where there is a broad consensus between the profession and the Revenue over how the disclosure regime will operate in practice. This consensus is almost certainly not strictly in accordance with what the legislation and the regulations actually say, but both sides, for pragmatic reasons, have been content to adopt a common-sense approach. At some stage I would expect that this consensus will be tested and it may well be that a more rigorous approach to the regime will have be adopted. In many ways this would be regrettable, because what we have at the moment does seem to work.

The key issue for the generalist adviser is to understand how the premium fee test works: it is this which turns what would otherwise be a reporting nightmare into a something which most practices are able to manage. I think that it is unlikely that an adviser will stray into premium fee territory accidentally. So provided that you are alive to the fact that if you do come up with a novel twist which you can exploit in the market place you will be within the disclosure regime, day-to-day life can probably continue as it always did.

And for that, we breathe a big sigh of relief.

ANDREW HUBBARD
Tax Director, Tenon Group

This article was first published in Tolley’s Practical Tax on 8 April 2005, and is reproduced with the kind permission of LexisNexis Butterworths