Tax return stats

The following quotes are worthy of note, either as a source of information, amusement or incredulity. You decide which. They are taken from the HMRC Consultation document
Meeting the obligations to file returns and pay tax, published on 19 June 2008 as part of the wider exercise: Modernising Powers, Deterrents and Safeguards. In most cases I have added my own obseravtions.

Para
1.3 - The first substantial measure, legislated in Finance Act 2007, was a single new penalty regime for incorrect returns for income tax, corporation tax, Pay As You Earn (PAYE), national insurance contributions (NICs) and value added tax (VAT) (the main taxes). [NB: this includes NICs which the Government refuses to accept is a tax]

2.3 - In 2006/07, HMRC expected to receive almost 23 million returns across all of the taxes. Of these, some 3 million were filed late or not at all. [That's an unlucky 13%. And seems very high so 'something needs to be done'. Ok, some returns will have been filed late and some may still be outstanding but how many were never due at all? I suspect that the 23m figure is the aggregate of all notifications to file that were issued by the computer. Looking back to prior years, what proportion of issued notifications have not resulted in a filed return after 3 or 4 years? New penalties (or incentives) are not going to make a difference here. The real figure of late returns is probably much lower.]

2.6 - It is important that HMRC receives a return even when there is no tax due. This is because HMRC needs to check the basis of the calculation by which the taxpayer calculated a zero liability. [Hmmm. Self assessment has always been described as meaning 'process now, check later'. Except that we know that HMRC do not (indeed they probably cannot) check all returns. So what are they saying here, if you have a nil liability your return will always be checked? Surely not.]

2.15 - The vast majority of people in the UK pay the right tax at the right time because they believe it is the right thing to do and because they are aware of the link between tax and public services. [Hmmm. I think the reason that most people pay the right amount of tax is fear of the consequences of not doing so. And one reason why tax compliance statistics have worsened since the introduction of self assessment is that fewer people are getting caught - so there are fewer people telling their 'war stories'. The direct consequence of this is that more people think they can get away with non compliance for longer.

3.4 - Despite the continuing efforts of HMRC to shorten and/or simplify returns, they can still appear complex and difficult to understand. ["Appear"??]

5.3 - It is important that before considering how any new penalty structure for late or non filing and payment could work time is taken to ensure that appropriate safeguards are in place to protect the taxpayer. [Hooray. The message is getting through]

6.6 For 2006/07, HMRC issued over 1.7 million penalties for late or non filing of an ITSA return. Of these some 153,000 were cancelled. This could be for a number of reasons, the most common of which is that the taxpayer successfully appealed against the imposition of the penalty. [Hmmm. According to para 6.8: 'of the 1.7 million penalties issued in 2006/07, 917,000 (over half) were capped.' - ie: not payable as the amount of outstanding tax was wither nilo or less than the £100 penalty. This suggests that almost 20% of the uncapped notices were cancelled (as distinct from the stated figure of 9% given in para 6.20).
That seems remarkably high, even more so when we are told that the most common reason was successful 'reasonable excuse' appeals.

7.2 - The available research and evidence suggests that the structure of a penalty is far more important than the level at which it is set in determining its effect on behaviour. [How strange. If the late filing penalty for self assessment tax returns was £500 rather than £100, I'm sure that most people would make more effort to ensure they file their tax returns in good time.]

7.13 - The suggestion made in consultation responses was for a' failure to notify penalty' to be suspended on condition that payment and filing obligations were met on time for a set period. [Yup. This makes sense to me too. Why then does HMRC continue in this paragraph to merely set out the difficulties that it sees with such a solution? Incidentally these are all expressed in terms of consequences that 'may' happen.]

7.34 - Assessments and determinations have the advantage of beginning the process of bringing something generally approximating to the right amount of tax into charge. [Wow. HMRC have crystal balls. I don't buy it. Where a tax return is outstanding HMRC will guess a figure that may be reasonable. It could just as easily be a gross underestimate or significantly more than the sum actually due. Indeed at para 7.52 we get the following worrying statement: "In most cases if the estimate is too high, the taxpayer can submit a return to overturn the estimate." - Most???

7.54 - It is important to note that from 2008 there will be two filing dates for Self Assessment. Those filing on paper will have to do so by the 31st October whilst those filing online can do so by the 31st January of the following year. There will therefore need to be two penalty dates – one for submitting a paper return after the 31st October and a second for failing to file a return at all by the 31st January. [As I predicted last year, the 31 October deadline is a red herring. There is no effective penalty regime in place for paper based tax returns filed between 1 November and 31 January.]

8.7 - There is a good argument that penalties should escalate for repeated failures, to discourage such behaviour. If this is appropriate, the challenge will be to combine this with recognition that prolonged failure should also be treated more seriously, and still produce a framework that is fair, proportionate and simple. [There is also a good argument that penalties should be remitted in the event of subsequent compliant behaviour - so as to encourage such behaviour. I can see no recognition of this 'carrot' approach as distinct from the 'sticks' set out in this con doc.]