A Proposed UK Code of Practice on Taxation for Banks – Abiding by the "Spirit of the Law" or "Spooky Jurisprudence" by another name?

July 07, 2009

On 29 June 2009, HM Revenue & Customs (“HMRC”) published a 24-page consultation document setting out proposals for a Code of Practice on Taxation for Banks (the “Code”).1  The consultation document outlines a “new and innovative approach”2  by the UK Government and HMRC to combating tax avoidance by banking groups operating in the UK.  The intention of HMRC, by introducing the Code, is to initiate and entrench changes in the behaviour and attitude of the banking sector towards tax avoidance.  The first paragraph of the proposed Code, which is currently in draft form, clearly articulates the expectation of the UK Government:

The Government expects that banking groups, their subsidiaries, and their branches operating in the UK, will comply with the spirit, as well as the letter of the law, discerning and following the intentions of Parliament3

The intention of HMRC is that the Code will engender a new level of transparency in the activities of the UK banking sector.  While HMRC’s expectation appears to be that all banking groups operating in the UK will sign up to the Code, the sanction for non-adoption will be to attribute direct reputational risk to any that do not.4  The Code sets out why the Government consider that a code of conduct is required, together with practical aspects of the administration and implementation of the proposed measures.  Comments on the draft Code, which is intended by HMRC to be the starting point in an ongoing dialogue with the banking sector and interested parties, have been invited by HMRC by 25 September 2009.

The context of the Code’s proposed introduction

HMRC identify that the context for the Code is an environment in which they consider that “more needs to be done” regarding tax avoidance schemes and transactions.

The Code refers to the current legislative environment in which Part 7 of Finance Act (“FA”) 2004 requires the early disclosure of certain tax avoidance schemes and to the corrective legislation which is regularly enacted to counteract tax avoidance arrangements.  Reference is also made to the commitments made in the London Summit of G20 leaders in April 2009 regarding the adoption by all jurisdictions of international standards for the exchange of tax information.  Nevertheless, despite an increasingly tighter legislative framework in which transactions motivated by tax avoidance are heavily targeted by HMRC, the Code expresses a concern that banks are uniquely placed to avoid their own tax liabilities or to provide services to their customers “which are sensitive to tax and some of which can be used for tax avoidance”.5  HMRC’s stated perception that the “unique position of the banking sector imposes a particular responsibility on them to comply with the spirit as well as the letter of the law”6 has therefore led to the Code under which HMRC wishes to adopt a new approach to tax avoidance.

The Code takes the current legislative approach to tax avoidance in a new direction.  The “new and innovative approach” is identified as being “one step further” than the current legislative environment in that it seeks not to counteract transactions but rather to modify behaviour and embed a series of responses and attitudes to tax avoidance within the banking sector.

It is considered that the approach set out in the Code does represent a fresh approach to tax avoidance, at least as regards the method by which HMRC’s aim is to be achieved, if not the aim itself.  The UK has no general statutory provision under which tax saving schemes can be void or recharacterised7.  In place of such a general statutory provision, Parliament has enacted anti-avoidance legislation targeting specific transactions and arrangements.  Furthermore, the English Courts have not evolved a jurisprudence under which transactions designed to avoid tax and otherwise lacking in commercial reality can automatically be rendered void or have their legal form disregarded.  At present, the Courts must have regard to the purpose of a particular statutory provision and interpret the statutory language in a way which best gives effect to that purpose8.  There is no explicit basis for Courts to tax transactions by reference to economic equivalence rather than their legal form unless the relevant statute expressly permits this.

The innovation of the Code is that it seeks to achieve a result through non-statutory means which is more normally achieved through legislation.  The Code would be a non-statutory and voluntary means to compel, through reputational sanction, certain behaviour amongst banks.  HMRC are open about this, stating that “[t]he Code does not replace any existing approach to tax avoidance and does not mean that the Government will not act quickly to shut down avoidance activity at the earliest opportunity”.9  However, the Code is less open about why HMRC might prefer a non-statutory approach in this area, and it is difficult to avoid the suspicion that HMRC prefer a non-statutory approach because it avoids the usual parliamentary process of implementation and will not be subject to interpretation by the Courts.  Difficult questions inevitably arise with the non-statutory nature of the Code, including whether HMRC will be exercising a public function in a sense which renders their decisions under the Code susceptible to judicial review.

Key components of the Code

The draft Code states that “banks” should:

a)      adopt adequate governance to control the types of transactions they enter into;

b)     not undertake tax planning that aims to achieve a tax result that is contrary to the intentions of Parliament;

c)      comply fully with all their tax obligations; and

d)     maintain a transparent relationship with HMRC.

These components are considered in greater detail in the remainder of this Memorandum.

These key components are intended by HMRC to deliver the principal objective of the Code, namely that banking groups will comply with “the spirit, as well as the letter, of tax law, discerning and following the intentions of Parliament”.10

In order to follow the Code the responsibility for discerning the “spirit” of tax law and the “intentions of Parliament” will rest with the bank concerned.  HMRC acknowledges that in “some circumstances” this process will not be straightforward.  The proposal from HMRC is that areas of uncertainty should be discussed with HMRC and notes that “HMRC will be happy to discuss the transaction and its understanding of the intentions of Parliament in relation to the law in question”.11

To some extent, the focus on the “spirit” of tax legislation is nothing new.  “Spirit” in the context of the Code might be construed as being the purpose which Parliament had in enacting the statute in question.  As Lord Hoffman stated in Norglen Ltd (in liquidation) v Reed Rains Prudential Ltd12:

“The question is simply whether, upon its true construction, the statute applies to the transaction.  Tax avoidance schemes are perhaps the best example.  They either work or they do not.  If they do not work, the reason....is simply that upon the true construction of the statute, the transaction which was designed to avoid the charge to tax actually comes within it.  It is not that the statute has a penumbral spirit which strikes down devices or stratagems designed to avoid its terms or exploit its loopholes.  There is no need for such spooky jurisprudence”.

The critical change proposed under the Code is that whereas statute is approved by the legislature (namely, Parliament) and then construed where necessary by the judiciary (namely, the Courts), the executive (acting through HMRC) will be the initial arbiter of whether a bank’s discernment of the “spirit” of the legislation under the Code is correct.  Another question is what the “spirit of the legislation” actually means.  In particular, the final form of legislation will often reflect compromises and significant digression from what the Government originally proposed.  In other words, whilst discerning the intention of the Government in proposing the legislation may be relatively easy, discerning the “spirit” or the intention of Parliament in enacting the final form of the legislation may be less so.  Indeed, that is presumably why judges have previously stressed that the true intention of legislation can only be discerned from its text.13

In any event, it is unlikely that HMRC will be able to oust the jurisdiction of the Courts absolutely.  It is at least possible that disputes between banks and HMRC as to whether there has been a breach of the Code will be litigated.  The resulting jurisprudence will be of great interest, as Courts may be expected to determine whether there is any real difference between the “intention” and the “spirit” of the law and what, if any, obligations arise on a bank to follow such “spirit”.  The cause of action which might be pursued by a bank is also an open question.  As mentioned above, the question of whether Courts will exercise jurisdiction over decisions of HMRC under the Code will depend on whether HMRC is exercising a “public function”.  The threat of reputational consequences for banks also evidences an intent to damage the reputation of the banks alleged to have breached the Code and raises the question of whether, in extreme situations, banks might contemplate bringing claims in defamation against HMRC and its officers where a published finding that a bank has breached the Code amounts to an untrue statement of fact (as determined by a Court).

Scope of the Code

The consultation document states the intention that the Code will apply to all banks operating in the UK, including foreign banks, together with “similar organisations undertaking banking activities”.  There is no clarification of what “banking activities” includes.  The Code therefore appears to encompass banking groups, their subsidiaries, “other vehicles” and banking branches operating in the UK.  The consultation document does not only specify entities which are regulated to do banking business by the FSA.  The net being drawn by the Code might well be wider, therefore, raising questions for a number of funds and even firms of financial advisers who might be regarded as providing banking services.


The Code will require that banks should have a documented strategy and governance process for tax matters encompassed within a formal compliance policy, including a documented strategy for complying with tax obligations.  The UK board of directors of the banks, or in the case of UK branches of overseas banks, a senior accounting person will be “accountable”14 for governance being compliant with the Code.

The only commitment that such a compliance policy is required to contain is a commitment “to comply with tax obligations and maintain an open, professional and transparent relationship with HMRC”.  The Code requires that “the policy must be taken into account in business decision making”.  Given that any tax sensitive transaction will invariably be reviewed by a bank’s tax department and credit committee, the circumstances where a tax policy is not taken into account as a matter of process will probably be limited in practice.  It is also likely that banks will already have a detailed strategy for complying with their tax obligations, with the risk of censure under the Code being be the least of their concerns if such compliance is not undertaken.  Accordingly, the governance requirements appear to be a less controversial aspect of the proposed Code.

Tax planning

The draft Code states that a “bank should not engage in tax planning other than that which supports genuine commercial activity”. The question arises as to what the word “genuine” adds.  The words of Lord Upjohn in IRC v Brebner15 shed some light on how far a bank might go and still be within the bounds of “genuine commercial activity”:

My Lords, I would only conclude my judgment by saying, when the question of carrying out a genuine commercial transaction, as this was, is considered, the fact that there are two ways of carrying it out - one by paying the maximum amount of tax, the other by paying no, or much less, tax - it would be quite wrong as a necessary consequence to draw the inference that in adopting the latter course one of the main objects is, for the purposes of the section, avoidance of tax. No commercial man in his senses is going to carry out commercial transactions except upon the footing of paying the smallest amount of tax involved.

In a Brebner sense, then, “tax planning ... which supports genuine commercial activity” (which would be permitted by the Code) arguably encompasses tax planning which seeks to eliminate any tax from applying to a genuine commercial activity.  In other words, there must be something other than tax which motivated the transaction and the tax planning must come second.  If “genuine commercial activity” can be viewed in this light, the Code may actually permit tax planning in a wide range of circumstances.

The draft Code goes on to provide that a bank “should reasonably believe that the transaction is structured in a way that gives a tax result which is not contrary to the intentions of Parliament”.  A similar expression is used both regarding where the bank acts as principal, and where it acts as the provider or facilitator of transactions by third parties.  Banks acting prudently and rationally and properly advised are unlikely to be acting in such a way as to have a reasonable belief that their actions are contrary to Parliament’s intentions (the discernment of which is considered above).  The difference with regard to the Code is that it is the Government (through HMRC) who want to be the arbiters of what Parliament intended as opposed to leaving that ultimate determination to the Courts where dispute arises.

There is also a reference to the bank acting as employer and the prevailing Government policy that all employees and employers should pay the “right amount of tax and national insurance” in respect of employment remuneration and rewards.  The position of the Government in this regard seems very similar to statements by the Paymaster General in December 2004 regarding tax planning in relation to employee remuneration.  The examples of tax avoidance in this area include “payments by way of assets” (potentially a reference to a number of exotic and, in many cases, long extinct methods of remunerating key employees by commodities outside the scope of the tax legislation such as platinum sponge and oriental carpets).

In this regard, the Code includes examples of what tax avoidance “can involve”, by using subjective concepts, namely:

a)      exploiting loopholes in tax law,

b)     juxtaposing two unrelated provisions in tax law in a way never intended or envisaged at the time the provisions were enacted; or

c)      artificially creating the conditions for a tax relief or deferral.

A number of hallmarks of avoidance are also listed in the Code including the use of offshore vehicles, arbitrage between tax jurisdictions using hybrid entities and instruments and “other arrangements, the purposes of which are to achieve a tax advantage”.16   These hallmarks are supported by a number of specific examples of aggressive avoidance. The body of the consultation document also lists further “signposts” of avoidance including transactions bearing little or no pre-tax profit, transactions which rely on mismatches (such as between legal and accounting form) or between the tax treatment of parties or entities in different tax jurisdictions.  A number of transactions bearing these “signposts” have been considered in the English Courts, but without such “signposts” being universally identified as being indicative of tax avoidance.17

There is ample scope for disagreement between the Government and banks on which legislative mis-alignments amount to loopholes, what was intended or envisaged by tax legislation or what constitute “contrived, artificial, transitory, pre-ordained or commercially unnecessary steps” in a transaction.18  Both the hallmarks and “signposts” identified in the consultation document and the proposed Code are indicative and not definitive.  This leaves open the question as to how the Government and banks will even discover that they disagree on such concepts.  The answer, presumably, is the concept of “transparency” which the Code implies is the solution to this problem as discussed below.

Relationship between banks and HMRC

The draft Code is expressed to be voluntary.  Signing up to the Code will bring certain obligations, one of which is the requirement that “relationships with HMRC should be transparent and constructive, based on mutual trust wherever possible”.19  The expectation in the Code is that where a bank “believes its proposed transaction may be contrary to the intentions of Parliament, the bank will explain its plans in advance with HMRC”.  There is no conditionality regarding this requirement: a bank signing up to the Code will therefore be required to engage in discussions with HMRC.  The threshold of this requirement is deliberately set low, being based on the bank’s subjective belief that a risk exists.  It is anticipated that considerable discussion will almost inevitably take place in the consultation process relating to the proposed Code regarding the nature of such a “belief”, and regarding the evidence which will be necessary to demonstrate that such a belief exists or does not exist.

Of the four parts of the proposed Code, it is considered that the proposals for the management of the relationship between HMRC and the banks under the Code may produce the greatest uncertainty in practice.  Of particular uncertainty at this stage is the focus on “transparency” as a key component of the relationship.  It is well understood that no tax planning or transactions should be entered into if they depend in any way on non-disclosure of information to HMRC.  However, the Code also anticipates the banks “disclosing fully the significant uncertainties in relation to tax matters” and disclosing “issues that HMRC would want to know about and might want to discuss”.20  The proposed onus on the banks of disclosing not only difficulties of interpretation which they have expressed but also issues which HMRC “might want to discuss” places a materially greater obligation on that bank than the routine self-assessment of their own tax liabilities.  The disclosure requirement under the Code of these matters appears to be without regard to how likely it is that the transactions in question will actually take place.  Amendment of the proposed code in this particular regard will no doubt need to be made to prevent time being wasted; a Code which requires the disclosure of sensitive issues in relation to a transaction which is very unlikely to take place seems to stand little chance of being workable in practice.

This onus on transparent disclosure also falls on a defined individual within each bank.  The consultation document proposes that the Code should be signed by a board member of the bank or a senior officer of the banking organisation.  HMRC propose raising any concerns over non-compliance by the bank with the signing board member or senior officer.  There is also the proposal that if a bank deliberately fails to comply with the Code, HMRC may make a report to any professional body of which that signing director or senior officer is a member.  This focus on individual responsibility for corporate action is another new feature of the proposed Code, although comparable changes are proposed in Finance Bill 2009 regarding senior accounting officers of large companies.21  Despite many banking organisations following best practice in their relationship with HMRC, this imposition of personal responsibility for the bank’s tax compliance arrangements is unlikely to be welcome.

For banks which are unwilling to sign the Code, the consultation document makes it clear that greater scrutiny from HMRC will be the result. While HMRC state in the Impact Assessment accompanying the consultation document  that “all banks will be expected to sign up to this”, a refusal to sign the Code will not, in itself, directly result in any additional tax liability.  Nevertheless, the inference to be drawn from the consultation documents and draft Code is that “direct reputational risk” could be occasioned by such a refusal. It is not particularly clear from the consultation document what form this “direct reputational risk” would take. It is not impossible that this could ultimately involve the publication of the names of banks who have not signed the Code with all the uncertainty that such a publication could create in the market and the possible adverse effect that such a publication might have to their share prices.


As mentioned above, HMRC appear to consider that certain key components of the draft Code, in their focus on adherence to the “spirit” of the law, might be seen as restating in a slightly different format existing canons of statutory construction as applicable to tax law.  (Less charitable readers of the draft Code may fear that HMRC’s construction of the “spirit” of the law may diverge from that of the House of Lords.)  It is clear that the Code will also remain voluntary.  Accordingly, and setting aside the expansion of information which HMRC anticipate should be disclosed as part of a “transparent” relationship, it is clear that HMRC anticipate that banks with good relationships with HMRC might notice little additional burden (apart from the initial implementation).22  Realistically, in this light, it must be asked whether the draft Code would have any real effect other than the “moral censure” of those who chose not to sign up.  The answer lies possibly in the Government’s own perception.  If the Government perceives that “inappropriate behaviour” persists, it states that it may introduce reporting or auditing requirements which examine a bank’s “compliance” with the Code and, it might be inferred, the mandatory adoption of the Code by banks.23

Far greater may be the UK banking sector’s likely concern over the intended change in attitudes and behaviours towards avoidance which is the aim of the Government under the Code since changes in this area may increase the competitive advantages of banks not following the Code.  Ignoring the possibility of UK banks not signing up to the Code and thereby deriving competitive advantages, there remains the possibility that foreign banks not lending from their UK branches (but rather, say, from elsewhere in the EEA) could stand at a competitive advantage in offering cheaper finance to UK corporates and businesses, such cheaper finance being enhanced through aggressive tax planning outside the UK.  Given the disparity of tax regimes even with the EU and the mobility of finance in the current markets, the Code’s possible creation of such competitive disadvantages may be a significant concern in the UK banking sector.

The real question is likely to be whether the Government, as the executive, genuinely believes that it can go any further without resorting to legislation as an agent of tax policy.  In this regard, the Government might do worse than have the words of Lord Wilberforce from Black-Clawson International Ltd v Papierwerke Waldhof-Aschaffenburg AG24 in mind:

Legislation in England is passed by Parliament, and put in the form of written words. This legislation is given legal effect on subjects by virtue of judicial decision, and it is the function of the courts to say what the application of words to particular cases or particular individuals is. This power, which has been devolved on the judges from the earliest times, is an essential part of the constitutional process by which subjects are brought under the rule of law – as distinct from the rule of the King or the rule of Parliament; and it would be a degradation of that process if the courts were to be merely a reflecting mirror of what some other interpretation agency might say …”.

It may be that the Government would accept that the imposition of more wide-ranging obligations upon banks and other tax payers than is currently envisaged by the Code would be a degradation of the constitutional process or simply unworkable in practice.  However, the concern must be that the Code presages the type of penumbral spirit or spooky jurisprudence which Lord Hoffman objected to in the Reeds Rain Prudential case referred to above albeit through extra-legal initiatives.

1   A Code of Practice on Taxation for Banks, Consultation Document, published by HMRC on June 29, 2009.

2   Consultation Document, paragraph 4.1

3   Consultation Document, Appendix 1, para 1.

4   Paragraph 35 of the UK Government’s document “Building Britain’s Future”, published in tandem with the Code on 29 June 2009.

5   Consultation Document, paragraph 2.10.

6   Consultation Document, paragraph 2.9.

7   While the UK tax statutes contain numerous examples of anti-avoidance provisions, frequently based around a tax avoidance “purpose test” or “motive test”, there is no all embracing single provision equivalent to Part IVA of the Australian Income Tax Assessment Act 1936.  While a general anti-avoidance rule (or “GAAR”) was contemplated by the UK Government in the late 1990s, the proposals were not legislated.

8   IRC v McGuckian [1997] STC 908 at 916 and Barclays Mercantile Business Finance Limited v Mawson [2005] STC 1 at 11.

9   Consultation Document, paragraph 3.5.

10 Consultation Document, Appendix 1: “The Code of Practice on Taxation”, paragraph 1.

11 Consultation Document, paragraph 3.22.

12 Norglen Ltd (in liquidation) v Reed Rains Prudential Ltd [1999] 2 AC 1 at 14, cited as recently as 25 June by the Court of Appeal in Prudential plc v Commissioners for HMRC [2009] EWCA Civ 622 at paragraph 13.

13 See for example Lord Hoffman, writing extra-judicially in 2005: “There is only one way to know the intention of Parliament and that is to read the statute”.  (BTR 2005 2, 197-206 at 204).

14 Consultation Document, paragraph 3.11.

15  IRC v Brebner (1966) 43 TC 705, at 718-719.

16 It is interesting that the consultation document invariably construes tax arbitrage between jurisdictions in a pejorative context and without benefit to the UK.  There is no consideration of transactions which might be initiated by, and generate economic returns for, UK banks but where the tax avoidance is the avoidance of the tax charged by a foreign jurisdiction, such through foreign tax credit utilisation structures.

17 For example, the Privy Council did not determine that the film financing arrangements in Peterson v CIR [2005] STC 448 were motivated by tax avoidance (which was proscribed by the relevant New Zealand anti-avoidance legislation) despite the pre-tax negative result of the arrangements and the circularity of the film funding arrangements.  The words of Lord Wilberforce in IRC v Holmden [1968] AC 685 at 712 are interesting in the context of “loopholes”:  “There is a presumption in taxing law that two sections, however complementary they appear, are exhaustive: there may always be a no man’s land between them which the subject does not have to define but on which he can take his stand”.

18 Consultation Document, paragraph 3.21.

19 Consultation Document, paragraph 3.23

20 Consultation Document, paragraph 3.24. The proposal that “any significant uncertainties” should be disclosed also appears, at least at first sight, being almost incapable of objective verification.

21 Clause 92 and Schedule 46, Finance Bill 2009 (as published on 30 April 2009).

22 Consultation Document, paragraph 4.9

23 Consultation Document, paragraph 4.10 and p.22.

24  Black-Clawson International Ltd v Papierwerke Waldhof-Aschaffenburg AG [1975] AC 591 at 629F-G.