Wittingly or no, HMRC is responsible for propagating misinformation on the new Allowances – the rot is spreading. But that may not be the worst of it.
AccountingWeb has just published a guide to the new so-called Dividend and Savings Allowances, which is available for download from their site. Regular readers will know that I have previously criticised the legislation in its draft form. Unfortunately, I have not had time to offer up a review of the significantly revised form in FA 2016 – I am under the cosh with other self-inflicted labours, which will, hopefully, bear fruit in the not-too-distant. Suffice it to say, for now, that the legislation is reasonably improved. The intersection between the, ahem, “Allowances” and the Transferable Tax Allowance (Marriage Allowance) is still a mess, which the aforementioned report does well to skirt.
I am, however, moved to write about one or two of the other points raised in the report. Because this really is getting out of hand.
Firstly, I have some sympathy for anyone trying to fashion a guide on simply how to allocate the Personal Allowance, in what is – let’s not forget – the current tax year, and already on the wane.
The author correctly notes that, now, the rule of thumb is that there is no longer a rule of thumb. But we actually have to go back before the allocation of the Personal Allowance: ITA 2007 s 23 Step 2 (loss reliefs, etc.) is also subject to the rule at ITA 2007 s 25 (b) that requires reliefs (and allowances) to be deducted in the way that will result in the greatest reduction in the taxpayer’s liability to Income Tax.
To give one a sense of the level of complexity now involved, here is the rule for allocating losses against dividends in the 2015/16 version of my tax calculator:
It doesn’t cover everything, but it’s reasonably accurate.
And now the 2016/17 version:
=MAX(0,InF+INFDAS+MIN(0,INDIIINet+InDIIIDAS+InDIIIGross+InTrIncome+InPAYEInc+InBIK+InPensionInc+InA-InRAC-InLossesetc))-IF(CUBRDivRte>CUDivTCRte,MAX(0,MIN (INDIIINet+InDIIIDAS+InDIIIGross+MIN(0,InTrIncome+InPAYEInc+InBIK+InPensionInc+InA-InRAC-InLossesetc),((InRAC+InLossesetc)-(InTrIncome+InPAYEInc+InBIK+InPensionInc+InA)),(MAX(0,(InRAC+InLossesetc)-(InTrIncome+InPAYEInc+InBIK+InPensionInc+InA))+(AdjPABeforeRestrToIncome+CUSavStartingRteBand+NewSavAllApplicable)-(INDIIINet+InDIIIDAS+InDIIIGross)),InFGross-MAX(0…. a few more lines to go yet, but you get the picture: hugely more complex.
As I have said before, I cannot wait to see HMRC’s 2016/17 Tax Calculation Worksheet. Proper programmers will see that there are duplications in the equation, but there is a trade-off between length and depth. (I am unabashedly on the shallow side).
I have not yet had time to check whether or not HMRC has issued its template for 2016/17 calculations, but if it’s anything like the fiasco with Class II NIC / the TTA and payments on account, then we are in for a very long tax return season next year, what with software houses being practically obliged to follow an HMRC-prescribed template that is patently flawed. Since I contributed to that thread, I have seen first-hand that HMRC's calculations do indeed (correctly) restrict the PoAs for the TTA. But software houses cannot, even if they know to, for fear of losing accreditation. Does HMRC do it deliberately to try to make us look stupid?
Anyway, back to the report:
“The value of the savings allowance is determined by the taxpayer’s adjusted net income, which is the taxpayer’s total net income after deduction of the personal allowance.”
Firstly, the value of the Savings Allowance is not determined by the taxpayer’s Adjusted Net Income. The value of the Savings Allowance is determined by whether or not the taxpayer is deemed to have “higher-rate income”, “additional-rate income”, or neither. (See the new ITA 2007 s 12B (3)). These terms are defined in the new legislation at ITA 2007 s 12B (8). It is broadly implicit that, if the new Savings Allowance were determined by Adjusted Net Income, then there would be no need for new, separate definitions. In fact, there is no reference to Adjusted Net Income in the new legislation – this should not really be a surprise to anyone, given ANI's purpose. There is reference in the new legislation to Step 3 of ITA 2007 s 23, which is fine. But Adjusted Net Income is defined at ITA 2007 s 58, and is primarily designed to determine entitlement to tax Allowances and more recently the HICBC, not exposure to Higher/Additional Rates of tax (etc) which is at ITA 2007 s 10…
If at this point you are thinking me the wrong side of pedantic, bar a smidge of trade union contributions and the like then OK: you say deduct pensions from income, I say extend the Basic Rate limit and arithmetically we’ll end up in pretty much the same place. But here is the rub:
“…adjusted net income, which is the taxpayer’s total net income after deduction of the personal allowance.”
Now that, dear reader, is cobblers. Adjusted Net Income is used to fix Allowances; Allowances are not deducted therefrom. But, hold on a cotton-picking minute, where have I read this particular brand of cobblers before..? Ah, yes, a few months back, when some poor soul on our forum had read .GOV guidance and assumed it had any substantive grip on reality.
It is perhaps worth recapping that, at the time, HMRC guidance explained Adjusted Net Income as follows:
“Add up your taxable income.
Include things like:
- money you earn from employment (including any benefits you get from your job)
- profits you make if you’re self-employed including from services you sell through websites or apps
- some state benefits
- most pensions (including the State Pension, company and personal pensions and retirement annuities )
- interest on savings and pensioners bonds
- dividends from company shares
- some rental income
- income from a trust
Take off any tax reliefs that apply like:
- payments made gross to pension schemes - those that have been made without tax relief
- trading losses, for example trade loss relief or property loss relief*
- personal savings allowance if you have savings interest and you are not an additional rate tax payer
- dividends allowance for part of your income if you are paid dividends...”
It seems the author of the report has simply followed HMRC’s guidance, in terms of the definition of Adjusted Net Income - which, if it were correct, might perhaps be a useful shortcut. The funny thing is, I have just checked the website and the .GOV guidance on Adjusted Net Income no longer says anything about the Savings or Dividend Allowances. They have finally updated the guidance as per my e-mails, after 3 months and 2 reminders – no, hang on a sec: according to HMRC's web page, this guidance was last updated on 4 February 2016… and yet no mention of those pesky Allowances… did I make it up? Am I looking at different guidance? Have I lost it? (Did I ever have it?)
No, I did not. Thanks to the power of the Internet, and in particular Rocky and Bullwinkle’s “Wayback Machine”, proof that I have not stepped into an alternate universe may be found at
So, to recap:
- HMRC’s guidance on Adjusted Net Income has been wrong since February 2016.
- It has clearly been relied on by members of the public for roughly the last 8 months, and at least one tax adviser, while wrong.
- HMRC has at last changed its guidance.
- But despite ostensibly having a system for recording changes to its guidance, it has not logged the update, so you would have no way of knowing the guidance had changed. In particular, if you as a tax adviser had based your work on that old guidance, you’d have no proof - nor reason not to doubt your sanity. (The shreds of mine remain intact, only by happy happenstance of having captured the original version in a previous post).
- The cat’s out of the bag, and the damage has now been done: the misinformation has found its way into further, non-HMRC guidance on which countless advisers will now rely, indefinitely.
It’s at this point that I am about ready to spit out my well-chewed dummy. I am so sick and tired of this… (you fill in the blanks) …that is continually being foisted on us. What is the point of supposedly cataloguing changes, when you don’t?
At the time of writing, I am due soon to meet one of the most capable and comprehensible tax practitioners I have ever met. It is supposed to be for a farewell coffee: a 21-gun salute minus the cannonade. I have already warned him that I have set myself the task of trying to convince him not to retire, because he would be a great loss and, right now, we need all the help we can get. Not just for me, mind, but for all of us and the taxpayers we advise. Then I reflect on this consummate drivel, and I can conclude only that I have, as a former colleague used to say, 4 hopes: