The Vodafone case—which, of course, is yet to be finally resolved—has triggered a debate in India about the legality and propriety of retrospective tax legislation. Much has been said about the latter but not very much about the former.
One issue that is often contested in relation to legality is whether the retrospective amendment is ‘clarificatory’. The reason this is important is not so much to ascertain whether the law should be construed to operate retrospectively—that question arises only if there is no express provision and one is seeking to rebut the presumption of prospectivity—but because it has a bearing on the validity of the law. The Supreme Court of India has called this the ‘principle of small repairs’ (the phrase appears to have its origins in an article in the Harvard Law Review). I discuss the cases on this subject in greater detail elsewhere ((2012) 5 SCC J-25)) but the underlying principle is essentially that it is easier to justify retrospective legislation if it simply confirms what should always have been the reasonable expectations of taxpayers. The best example (although not a tax case) is Zile Singh v State of Haryana. In that case, the State of Haryana passed a law seeking to disqualify those with more than two children from contesting certain elections but intended to provide a one-year amnesty. In the proviso granting the amnesty, the word ‘after’ was inadvertently used instead of the word ‘upto’ with the result a candidate was disqualified if he had more than one child until one year after the amendment but not thereafter. This was obviously not what the legislature had intended—and more importantly not what any candidate could reasonably have thought it had intended—and legislation correcting the error was construed to be retrospective by the Supreme Court. If it had been made expressly retrospective, a challenge to its constitutional validity would undoubtedly have failed.
Some of these issues—although in a different context—have been considered by the English High Court in its recent judgment in R (St Matthews (West Ltd)) v HMRC in which retrospective tax legislation (something of a rarity in the UK) was unsuccessfully challenged by the taxpayer. The Finance Act 2003 introduced what is now the ‘Stamp Duty Land Tax’ (‘SDLT’) regime in the United Kingdom. The essential scheme of SDLT is that a certain percentage of the consideration is payable as stamp duty on the sale of land used as a residence within the UK. When the Bill was being debated, a concern was expressed about potential double taxation if A enters into a contract to sell land to B, but B then (before completion) assigns the benefit of the contract to C, in whose favour A executes the conveyance. To avoid this risk, section 45 of FA 2003 provides that SDLT shall not be payable if there is an assignment, sub-sale and substantial performance of the original as well as the secondary contract.
The intention of s 45, as Mrs Justice Andrews notes at , ‘was to place the taxation burden on the person who is going to have the use and enjoyment of the property.’ But a number of ingenious tax avoidance schemes were devised to take advantage of this provision. One of these was called the Blackfriars scheme. Under this, A and B would exchange contracts for the sale of property at market value; B would agree to grant C (normally a connected person) an ‘option’ to purchase the property on the date when the original contract completes, for a price marginally higher than the SDLT threshold but at a fraction of the market value. Both agreements would be ‘substantially performed’—the first contract by completing and the second by B granting the option he had agreed to grant. It was said that s 45 therefore applied and that this transaction was outside the SDLT regime.
There were doubts about whether the Blackfriars scheme actually worked but the matter had not been tested when Parliament enacted retrospective amendments making it clear beyond doubt that the Blackfriars scheme did not work. The question was whether this retrospective amendment could be challenged as contrary to article 1 of Protocol 1 of the European Convention of Human Rights (‘A1P1’). A1P1 provides, in essence, that every person is entitled to the peaceful enjoyment of his possessions of which he may not be deprived ‘except in the public interest’. There was some doubt about whether A1P1 was engaged (see –) but the interest in the case for our purposes lies in Mrs Justice Andrews’ analysis of the position on the assumption that it was. The two requirements for validly depriving a person of his possessions are that the interference must be lawful and proportionate. Two factors, in particular, persuaded Mrs Justice Andrews that this retrospective amendment was clearly lawful and proportionate. First, a warning was given by the Chancellor of the Exchequer in 2012 that the Government would use retrospective legislation to invalidate any SDLT avoidance scheme in this specific context. Secondly, and especially in the light of that warning and measures taken in relation to other SDLT schemes, it could not be said that there was any legitimate expectation that this scheme would succeed:
65. In my judgment none of these arguments has any merit. In the wake of what was said by the Chancellor at the time of the 2012 Budget, any person who was well advised and who gave even cursory consideration to the issue must have appreciated that it was highly likely that once HMRC became aware of a variant on an existing tax avoidance scheme based on the transfer of rights rules in the FA 2003 which had been rendered ineffective as from the 2012 Budget, it would take swift action to put an end to the variant as from the same date.
It was also suggested that Parliament had not enacted similar retrospective measures to invalidate other SDLT avoidance schemes and that it was unlawful to attack just the Blackfriars scheme. This argument failed as well:
69. If other tax avoidance schemes unrelated to SDLT were not the subject of similar legislation it does not follow that there was anything arbitrary or capricious about this legislation or about its operation. Mr Beal’s riposte to that argument was that it is not open to the Claimants to seek to take advantage of an alleged failure by HMRC to apply the (same) correct tax treatment to someone else, because two wrongs do not make a right. I agree. Moreover since it is incumbent on Parliament to make decisions based on relevant facts and circumstances, no inferences can possibly be drawn from any decision not to make anti-avoidance legislation retrospective in unrelated areas, in which the relevant factors might well point towards a different conclusion being reached as to the proportionality of that approach.
The fundamental difference between Vodafone and cases of this kind—apart from the fact that Indian law arguably imposes greater restrictions on retrospective legislation than does English law—is that there was almost certainly a reasonable expectation that the transaction was not taxable: an expectation which the Supreme Court confirmed was based on an accurate analysis of the Income Tax Act, 1961, as it then stood. It is difficult to see how the principle of small repairs is attracted: the amendment appears to be neither repair nor small, notwithstanding its characterisation as ‘clarificatory’ in the Finance Act, 2012. But the case importantly illustrates that retrospectivity alone is not enough: much depends on exactly what the position was before the law was amended and whether a reasonable taxpayer would have thought that tax was not payable.