David Whiscombe of BKL Tax laments uncertainty in tax, and provides two unfortunate examples faced by taxpayers and advisers.

More uncertainty... 

David Whiscombe

Which brings us to two recent developments.

Dividends as employment income

The first is the publication of the updated HMRC guidance on the revised ITEPA s447.  This legislation deals with the taxation of “post-acquisition benefits from employment-related securities” and with the circumstances in which such benefits (and in particular dividends) are chargeable as employment income.

The scope of s 447 is literally doubly important: if dividends are chargeable under s 447 as employment income this is in addition to, not in substitution for, the normal charge to tax on them as dividends.  Strangely, the government seems to have overlooked telling MPs when the legislation was before the House that they were being invited to vote for double taxation: but the potential is explicitly recognised in the Employment Related Securities Manual at paragraph 90210.  So it is particularly important for taxpayers to know whether they are going to be caught by s 447.

When the revisions to s 447 were first proposed, doubts and concerns were expressed about their application to dividends, and the HMRC guidance hardly resolves them.  Quoting the comments of Dawn Primarolo (the Paymaster General whose previous jobs as bookkeeper, legal secretary and typesetter obviously make her eminently well-qualified to interpret some of the most complex tax legislation ever introduced) HMRC say that the rules will apply where there are “complex, contrived arrangements to avoid paying Income Tax on employment rewards” which are used to avoid paying “the proper amount of tax and National Insurance”.  And there lies the first difficulty: the “proper amount of tax”, Dawn, is conventionally the amount which Parliament has imposed by clear words – not the amount which you or some other functionary arbitrarily think a person ought to pay.  It is simply not good enough for Government to impose taxes in the most general and wide-ranging terms and then to assure us that the legislation will be applied only to collect the “proper amount of tax” – that amounts to taxation by the Executive and not by Parliament and it is just plain wrong.  Apart from any constitutional reservations the practical concern is that departmental pronouncements about the way in which in which it is intended that legislation should be applied can have all the enduring qualities of the Seven Commandments of Animal Farm…  For example, when the draconian “new” s20 of Taxes Management Act 1970 was being introduced in 1976 it was in the spirit that its use would be restricted to the most serious cases: a generation later its use has become almost routine.

The main problem of course is that the words “complex contrived arrangements” are Dawn’s gloss and they do not appear in s447: the legislation imposes an employment income charge on dividends where “something has been done which affects the employment-related securities” as part of a scheme or arrangement aimed at tax avoidance.  What on earth does that curious phrase mean?  Does simply paying a dividend “affect” the shares on which it is paid?  Does creating a new class of share “affect” that class?  The man on the Clapham omnibus would probably answer “no” to both questions.  HMRC disagree.  At ESRM 90210 it is baldly asserted that dividends on so-called “alphabet shares” issued to employees are within the s447 charge.  Of course there is no real doubt about the fact that the creation of such shares and the payment of dividends is part of a scheme or arrangement aimed at tax avoidance.  But “something done which affects the employment-related securities”?  HMRC must think so.

So what about two or three individuals who form a company and operate it as a quasi-partnership?  They will typically agree to share profits and pay themselves dividends according to their respective input and they may very well create for themselves different classes of share to facilitate payment of differential dividends.  Do HMRC tar that arrangement with the same brush as alphabet shares for employees?  If not, why not?  Is that not just as much “something done which affects the employment-related securities” as are alphabet shares for employees?  And what of the near-universal practice of the owner of a one-man company paying himself a small salary and voting most of the profit as dividend?  Plainly it’s done to avoid NIC.  But do HMRC think that s447 applies?  Unfortunately it’s impossible to discern an answer from the weasel words of the “guidance”.  Dawn offers the comfort that “where investors are carrying out their normal investment transaction, this charge will not affect them” – and that “this measure will not affect the taxation of those small businesses that do not use contrived schemes to disguise remuneration to avoid tax and National Insurance”.  But that begs the question – is establishing a one-man company a “normal investment transaction”?  And is paying yourself next to no salary and a large dividend a “contrived scheme”?

Uncertainty on residency

For our second recent example of uncertainty look at the Gaines-Cooper case.  It has been widely recognised for many years that the statute and case law surrounding questions of residency is in a state that is technically known as a pig’s breakfast.  Even before the ease of transglobal travel and communications and the rise of the global village made boundaries increasingly pervious to ideas, money and people it was already practically impossible to make any rational sense of the law; and the rise of international commuting has been the final nail in the coffin of any hope of getting much practical help from the (mostly nineteenth and early twentieth century) case law.  Thus the codification of HMRC practice set out in IR20 has assumed great practical importance and has become generally accepted as the rule book for such matters.  IR20 might be imperfect but at least one thought one knew where one stood.  IR20 says that the normal rule is that days of arrival and departure are ignored in counting the number of days of presence in the UK.  Mr Gaines-Cooper did not, on that basis, average more than 90 days a year in the UK: he therefore believed he was home (or rather, not home) and dry. 

IR20 does create anomalies: many will agree that it is odd that someone who spends in aggregate many days physically present in the UK should be non-resident simply because his presence includes a large number of short trips.  And perhaps there is an argument that IR20 should be revised.  But, note – “revised”: not “ignored when it seems like a good idea”.  Look, HMRC – you wrote the rules– you encouraged us to play by them – it’s frankly unacceptable for you to change them halfway through the game or – as in the case of Mr Gaines-Cooper, to decide towards the end of extra time that the rules have been changed with effect from the start of the game.

In Revenue & Customs Brief 01/07 HMRC make a valiant but misguided attempt to defend the indefensible and invite us to believe that the approach in Gaines-Cooper is wholly consistent with IR20.  If you have been resident at any time in the UK, say HMRC, you have first to decide whether you have left the UK before you move on to the stage of counting days of presence in the UK.   What was fatal to Mr Gaines-Cooper, they say, is not that his days of arrival and departure have to be counted: it’s that he never left the UK in the first place.  There are a number of objections to this.  First, it’s not the case that seems to have been put before the Special Commissioner – rather it’s an ex post facto attempt to square the circle.  Second, if this approach be true, it follows that two individuals whose actions and behaviours in a given tax year are absolutely identical may be treated differently for residency purposes depending upon whether they have at some time in the past lived in the UK – a surprising result to say the least.  Third, there is nary a word of this fine distinction in IR20.  On the contrary the very strong impression given by IR20 is that an individual who makes his home elsewhere and who habitually and regularly spends less than 91 days in the UK will be treated as ceasing to be resident and ordinarily resident.  Full stop.
 
There is already more than enough uncertainty in taxes.  We need more of it about as much as (to paraphrase John Donne) Argus needs more eyes, or women or the sea more tears.  So how about HMRC start seeing their job as promoting certainty rather than removing it?

Article Added Thursday, 01 March 2007 | 3857 Hits

Your attention is drawn to the disclaimer on this site, which applies to the content in this section.