Michael Thomas, Barrister, explains the scope and potential implications of the new Stamp Duty Land Tax anti-avoidance rule.
Section 75A, the new general anti-avoidance rule for SDLT, was first introduced by the SDLT (Variation of the FA 2003) Regulations 2006 (SI No. 3237) with effect from 6 December 2006. A slightly revised version (which will have retrospective effect to 6 December 2006) is being enacted as clause 70 of the 2007 Finance Bill and it is that version which forms the basis of this note.
The striking feature of s 75A is that it is very widely drafted. It is therefore a very effective weapon available to HM Revenue at Customs (“HMRC”) and, is likely to succeed in countering the SDLT planning arrangements against which it is designed to be used if it is deployed before the courts. To a large extent it is thought that s.75A will indeed mark the end of SDLT “schemes”. The introduction of s 75A will accelerate the trend towards bespoke planning which is of more limited application. The potential width of situations in which s 75A might apply gives cause for concern that it may catch innocent transactions. However, it is thought that difficulties should not arise provided that both HMRC and the courts give s.75A a sensible interpretation.
Conditions for s.75A to apply (see s 75A(1))
Section 75A adopts the approach of “following the land”. It applies where a vendor (“V”) disposes of a chargeable interest (i.e. land) and a purchaser (“P”) acquires either that interest or an interest deriving from it. A number of transactions must be “involved in connection with the disposal and acquisition”; these are referred to as “the scheme transactions”. The final condition for s.75A to apply is that the total SDLT payable in respect of the scheme transactions must be less than “the amount that would be payable on a
notional land transaction” under which P acquires V’s chargeable interest. The conditions for s 75A to apply will now be considered in more detail.
What is a scheme transaction?
Section 75A(2) and (3) contain further details on the kind of “scheme transactions” at which s 75A is aimed. “Transaction” expressly includes “a non-land transaction”, “an agreement, offer or undertaking not to take specified action”, “any kind of arrangement whether
or not it could otherwise be described as a transaction” and “a transaction which takes place after the acquisition by P of the chargeable interest” (see s 75A (2) (a) to (d)). Although the term “transaction” is not expressly defined it is apparent that it is to be construed very broadly and that s 75A(2) is designed to achieve this. An example of a non-land transaction is the sale of shares in a company which owns the relevant land.
A list of six examples of scheme transactions is provided in s 75A(3). The purpose of this is to ensure that both taxpayers and the courts are in no doubt that Parliament has stopped certain schemes in accordance with HMRC’s recommendations. Two of the examples given are a sub-sale and the carving out of a lease from a freehold. The remaining examples all relate to the right to terminate a lease and comprise the grant of such a right to terminate (whether as part of a new lease or by the variation of an existing lease), the exercise of such a right and an agreement not to exercise a right to terminate a lease or to take some other action.
“In Connection With”
The scheme transactions must take place “in connection” with both the disposal and the acquisition. The interpretation of this phrase is crucial to the scope of s 75A. The statute does not qualify the term “in connection with” and this suggests that it is intended to have the broad interpretation for which HMRC would no doubt contend. However, there are problems with having a concept which is so potentially broad at the heart of s 75A. One is that innocent transactions, where there is no question of the taxpayer trying to avoid SDLT, are potentially caught. It is suggested that if this issue does ever come to court then a sensible interpretation will be given to s 75A to ensure that it only catches the kind of “schemes” at which it is clearly aimed. At least, that is what should happen but there are no guarantees in litigation. The result is needless uncertainty, which is the second problem.
Uncertainty wastes the resources of taxpayers, whose advisers will have to consider the risk that s 75A might apply. Moreover, the uncertainty works against HMRC because it encourages those who design schemes to save SDLT to speculate what limits the courts might set on the relevant connection and to try and circumvent s 75A using that limitation. This is of course inherently a very dangerous exercise because when confronted with aggressive planning intended to circumvent a mini general anti-avoidance rule the likelihood is that the courts will find that no such limitation exists. However, if there are no other downsides to adopting the planning, then there may be taxpayers who are prepared to take the risk and make HMRC fight and win the point. For example, it might be argued that the relevant connection is lacking if the vendor is unaware of the scheme transactions which are to follow, such as where the first
step involves a “double completion sub-sale”. The problem for any taxpayer seeking to argue this is that there is nothing in s 75A to require all the parties to have a subjective intention for the scheme transactions to take place. All that is required is for the scheme transactions, looked at objectively, to take place in connection with one another. In practice the result is to a large extent a matter of impression: if a court finds the transactions to be the kind of planning which should be struck down then it is likely to find the requisite connection and if it regards the transactions as innocent then the opposite will follow.
It is much more strongly arguable that the requisite connection is lacking is if the scheme transactions are undertaken by the vendor before any purchaser arrives on the scene. SDLT is a purchaser-orientated tax and if the steps taken to save tax do not involve him, especially if there is a long gap before his involvement, then s 75A should not apply as the requisite connection will be lacking. To take a simple example, if V creates a special purpose vehicle company in advance of marketing that company and in the future the company is then marketed and ultimately acquired by P then s 75A should not apply. Indeed this appears to be expressly accepted by s 75C(1). Of course, creating an SPV is difficult owing to the anti-avoidance legislation aimed at preventing this. However, there are situations where it is possible, such as where other assets are transferred out of a company which also owns land in order to create an SPV (It might be argued that the stripping out of the other assets from the SPV does not qualify as a scheme transaction because it takes place before any land is disposed of. In any event, the present issue is when s 75A might apply) or if land is transferred to an unconnected company.
The Notional Land Transaction
The notional land transaction fulfils two functions. First, it forms a key part of the test to determine whether or not s 75A applies. Secondly, if s 75A does apply then the charge is computed by reference to it as discussed further below. As mentioned above, the notional transaction involves the acquisition of V’s chargeable interest by P. The consideration under the notional
transactions is the larger of the aggregate amounts either given by way of consideration by any one person for the scheme transactions or received by V or a person connected with him. Scheme transactions which are also land transactions are disregarded for the purposes of SDLT. The chargeable consideration on the notional transaction includes deemed chargeable consideration
arising under s 53 FA 2003, the SDLT partnerships regime and the rules on exchanges (see s 75C(5) and (7)).
The Effect of Section 75A Applying
When section 75A applies, then two consequences result. First, any scheme transactions which are also land transactions are disregarded for SDLT purposes. Secondly, tax is then charged on P by reference to the notional transaction, the effective date of which is the earlier of the last date of completion for the scheme transactions or the last date on which a contract in respect of the scheme transactions is substantially performed. The result is a kind of statutory Furniss v Dawson where REPLACE ed steps are disregarded and tax is charged according to the end result. However, on closer inspection, s 75A is almost certainly wider than Furniss, for example because of the width of the term “in connection with”.
Exceptions to Section 75A
The revised statutory wording in the Finance Bill provides for several circumstances when s 75A will not apply. However, what is most striking is the absence of any statutory motive test or clearance procedure for transactions which do not have tax avoidance as one of their main purposes. The author’s view, as stated above, is that the courts will interpret s.75A to achieve this result but uncertainty still remains. HMRC are similarly of the view that s.75A exists to prevent abusive avoidance but of course that is no comfort when a taxpayer and HMRC disagree as to what is abusive.
The statutory exceptions to s.75A will now be dealt with in turn.
First, s 75A does not apply where the SDLT payable in respect of the scheme transactions exceeds that payable on the actual transactions only by reason of either the reliefs for alternative financing arrangements under ss 71A to 73 or the social housing provisions contained in Schedule 9. This exception applies where tax is reduced “only” by the relevant provisions. Accordingly, it is not possible to have, for example, a prior sub-sale combined with an alternative finance relief provision as a basis for planning and then claim that s.75A does not apply.
Secondly, in calculating the chargeable consideration on the notional transaction the consideration for what would otherwise be a scheme transaction is ignored if it is “merely incidental” to the transfer of the land from V to P under s 75B(1). It is doubtful whether s 75B really adds anything because if a transaction is “merely incidental” to the land transfer, then it is highly arguable that it lacks the relevant connection with it in any event. Moreover, it is expressly provided (by s 75B(2)) that a transaction is not incidental if it “forms
part of a process, or series of transactions, by which the transfer is effected”, “the transfer of the chargeable interest is conditional on the completion of the transaction” or if it is of one of the specific scheme transactions listed in s 75A(3). So, it is difficult to imagine scenarios where the exception for incidental transactions will make a difference in practice. Nevertheless, it is provided that a transaction may be incidental if it is undertaken only for a purpose relating to the construction of a building, the sale or supply of something other than land or a loan to P to enable him to acquire the property (see s 75B(4)). As stated above, these kind of transactions lack the relevant connection with the land transfers in any event.
Thirdly, a transfer of shares is ignored for the purposes of s 75A if it would otherwise be the first in a series of scheme transactions (see s 75 C (1)). This ensures that a straightforward sale of a land owning company followed by a liquidation is not caught by s 75A; although it is doubtful whether it would have been caught in any event.
Fourthly, the notional transaction under s 75A attracts any “relief” as if it were an actual transaction. The term relief is not defined so that there may be some uncertainty as to what is a relief: for example strictly this would not cover any exemption within Schedule 3. More fundamentally, this does not assist in the situation where there is a series of innocent transactions, each eligible for individual reliefs for which the notional transaction would not qualify. For example, land might be transferred up to a Newco with group relief claimed prior to a liquidation reconstruction on which reconstruction relief is claimed. This arrangement cannot be intended to be caught by s 75A but the notional transaction would qualify for neither relief. An attempt seems to have been made to address this very point in s 75C(3) but it does not solve the problem because it merely provides that the notional transaction satisfies the statutory purpose test if
any of the scheme transactions do. This provision needs amending and this may happen during the passing of the Finance Act.
Fifthly, no account is taken of any consideration paid in respect of certain transactions which qualify for specified reliefs (see s 75C (4)). This means that even if the notional transaction does not of itself qualify for relief, then some or all of the consideration may not be chargeable.
Finally, on a practical note, HMRC has issued a so-called “White List” on transactions which will not be caught by s 75A. The examples given are clearly outside s 75A in any event, so the White List has little practical use. However, it is worth noting that Prudential planning (see below) involving separate sale and build contracts is expressly not caught.
What is Caught by s75A?
As stated at the outset, s 75A is a widely drafted mini general anti-avoidance rule. It catches several planning arrangements which were popular prior to December 2006. These include the more aggressive planning based on sub-sale relief under s 45(3) designed
to ensure that little or no tax was paid when there was no genuine commercial sub-sale, the various ideas based on terminating a lease to radically alter its value without an SDLT charge and the idea of the purchaser paying if the vendor failed to exercise some right, such as the right to terminate a lease. Schemes involving partnerships are also caught although the flaws in the legislation on
which the planning was typically based are dealt with in their own right, as discussed further below. Nevertheless, s 75A prevents individual vendors from using partnerships to avoid SDLT being payable on a sale.
Are Innocent Transactions Caught?
The short answer to this is that innocent transactions should not be caught. However, the position is not as clear as it could be owing to the width of the scope of s 75A. The result is that there is some uncertainty. However, in the author’s view very often it will not be too difficult to form a firm view that s 75A does not apply when there is no question of any planning.
What Planning Survives s 75A?
Perhaps the biggest question in practice is what planning survives s 75A. It is possible to identify various arrangements which clearly do.
One is a Prudential arrangement where the purchaser acquires bare land from a vendor and at the same time enters into a building contract with the vendor so that SDLT is payable on the land value alone.
Another is a sale of shares in a land owning company or units in a unit trust. Creating a special purpose vehicle for sale is a more difficult exercise of course. Prospective corporate vendors should consider creating SPVs more than three years in advance of any
prospective sale to avoid any clawback of relief. SPVs can also be created by hiving out other assets to create a clean company where the other factors permit this.
A third kind of arrangement which should be immune from s 75A is for a developer to not acquire any interest in land but instead to be paid as a builder and also as marketing agent for the landowner-vendor. Section 75A has nothing to bite on provided that the developer does not acquire any interest in land. The key to ensuring the success of this kind of arrangement is to ensure that the developer does not acquire any interest in land whilst balancing this against the commerciality of the deal and the vendor’s needs. These kind of
arrangements might well become very popular over the next few months.
A fourth group of arrangements comprise what might be termed “bespoke planning”. In some transactions there will be particular facts which mean that the SDLT charge can be reduced without the need for a contrived arrangement. Put another way, if there is no series of transactions, then s 75A cannot apply. One example of this is for a tenant to swap a 999 year lease for a new lease rather than a freehold to take advantage of the rules on surrenders and regrants. Another is the rule that no SDLT is chargeable on the incorporation of a partnership, as discussed further below.
Finally, there may be scope for more aggressive arrangements. However, the most likely key to achieving this is to break the "connection” test. It is considered that there will be good arguments that the connection test is broken if the vendor sets up the planning prior to the purchaser arriving on the scene. The longer the time gap between the setting up of an arrangement and an ultimate sale, the harder it is to argue that the requisite connection is present. The Furniss v Dawson case law is relevant in
this regard. If no consideration passes at the first stage then there is nothing for s 75A to bite on at that time. However, if there is to be a significant time gap, it must be recognised that in many cases it may not be worth all the effort, risk and commercial inconvenience in order to save SDLT at 4% and it might be simpler to create an SPV and wait 3 years. There may be scope for other arrangements which exploit the s 75A rules for the basis of planning, such as by taking advantage of the rules on deemed consideration to achieve an SDLT saving. One such arrangement along these lines is currently available but will be blocked when the Finance Bill receives Royal Assent.
HMRC’s hope is clearly that s 75A will mark the end of “one size fits all” SDLT planning schemes. To consider whether this is likely to be the case it is necessary to understand HMRC’s approach to SDLT planning. HMRC has up to now been slow to attack SDLT planning arrangements through the courts. Instead the approach has been to amend the legislation. Not all of the legislative amendments have been unqualified successes as more than one has failed to stop the intended target and others have been unnecessary and only caused further confusion. It is also important to understand that not every arrangement which s 75A
targets necessarily worked anyway. HMRC would have had strong arguments against more than one of the schemes against which s 75A is targeted. Who would have prevailed before the courts is inherently uncertain but it is very unlikely that HMRC would have failed in every challenge.
Conversely, HMRC’s failure to challenge SDLT schemes encourages taxpayers who wish to adopt them. If HMRC accepts that a scheme works then that is its result. If no other tax apart from SDLT is at stake then more aggressive taxpayers may be prepared to undertake planning on the basis that little will have been lost if it fails. Any statutory provision is open to interpretation and by using such broad charging concepts s 75A lends itself to possible interpretations which radically restrict its scope. In short, if HMRC wants to stop the next round of SDLT schemes, then it is likely that it will need to actually use s 75A. The author’s view is that s 75A will
indeed be used, in particular because there is no basis for HMRC to ask Parliament for anything wider!
So, whilst it may not be impossible to plan aggressively around s 75A that exercise will be both very difficult and very high risk. Before any arrangement is adopted then those risks will need to be spelt out in full to any client. Almost certainly full details will need to be
disclosed to HMRC to prevent disclosure assessments, penalties and any accusations of impropriety. Expert advice will need to be taken in individual issues.
Section 75A is a widely drafted and very powerful mini-general anti-avoidance rule. Provided that HMRC actually uses it, then it is likely to spell the end of most if not all “one size fits all” SDLT schemes. The future of SDLT planning is will be towards bespoke ideas and
planning of which HMRC approves. When undertaking it should not be overlooked that aside from s 75A it needs to work as a matter of the general SDLT code, not fall foul of Ramsay and fit with both other taxes and the commercial deal.
The above article was first published in 'GITC Review'(Volume VI Number 2, June 2007), and is reproduced with the kind permission of the author.
Michael Thomas is the author of the book 'Stamp Duty Land Tax', which is published by Cambridge University Press
With contributions by David Goy QC
Full details of the book are available at: http://www.cambridge.org/uk/catalogue/catalogue.asp?isbn=9780521606325