Dated November 27th 2007
Clive Lewis asserts that the council data available to the auditor will include
The information sought in this instance will contain the unique council tax reference
for an individual, a unique property reference number and the full name, address
and date of birth of individuals residing at the address and liable to pay council
tax, including information about whether or not the individual is claiming that he
or she is eligible to receive discounts from the full council tax normally payable.
No it won't.
First of all, it won't because the council will not have information about individuals residing at the address and liable to pay council tax. It may have it for some addresses: it will not have it for all addresses, and why should it?
Mr Lewis is unclear here about which 'discounts' (plural) he is referring to. In other parts of his opinion he uses the phrase 'sole occupant discount', a misleading term which does not appear anywhere in the law. When Mr Lewis uses the word 'but' in a context where more than one adult is on the electoral register this implies that he has been instructed that the persons in questions are claiming to be entitled to a 25% discount on the basis that they live alone. This belief is completely incorrect and it flies in the face of council tax discount law.
No individual whose data is used for this exercise is claiming to be eligible for 'discounts' (plural). The information does not include copies of any claims. The exercise relates to just one 'discount' - a discount of the appropriate amount under Section 11(1) of the Act, which has been awarded on the basis that the same amount of discount will apply on every day of the coming tax year.
One has to ask whether Mr Lewis came up with these ideas himself or whether it was part of his instructions.
For it is clear that way before the date of his opinion the NFI Audit Guide was getting the law wrong. And in doing this it blames, no, sorry, that should be 'thanks', Hillingdon, the incorrectness of whose practices needs no rehearsal here. They did the data mining in June 2006, and three more councils followed suit that October.
So the more you think about it, the more it seems more likely than not that Clive Lewis was issued with legally incorrect instructions and that, perhaps reasonably, perhaps not - given the degree of controversy, he did not bother to check them out. He was not asked for a briefing on how councils must by law adminster the discount, after all. He would not have been the first to take what the NFI said at face value: this almost appears to be something that the Audit Commission trades on.
More evidence for the belief that Clive Lewis was working with misleading instructions comes in the Audit Guide for the SPD roll out dated 2006/2007. This predates his opinion by some months.
It states that in certain cases selected for 'matching' there is no ambiguity about the reason for the discount, an utter piece of nonsense which the Audit Commission's own legal briefing has utterly demolished in a message which has still not yet got through to all Audit Commission staff. There is no legal requirement for the council to know on a day to day basis the reasons why the discount applies, and no legal power for the Audit Commission to use data matching to force councils to carry out their duties under Regulation 14 in particular ways.
Dubious statistics
Clive Lewis appears not to be a statistician or to have checked what the 'pilots' actually showed in terms of money. If he describes his 'instructions' accurately, they were themselves misleading and biased.
This information is being required in the context of the National Fraud Initiative.
In brief, pilot schemes in London have demonstrated that a large amount of
revenue (in the region of £50-100 million pounds annually) is being lost by local
authorities.
In brief, pilot schemes have demonstrated no such thing.
The actual claim made in briefing note dated April 2007, and signed by Peter Yetzes, was that using particular methods, you could categorise very large numbers of discounts as 'questionable', and on this basis the NFI make a guess about how much money nationally 'could' be being lost to fraud.
The figure given by Mr Yetzes at that time (2007) was over 100 million nationally.
And it is clear that Mr Yetzes was working on the basis of false beliefs about how council tax discounts work.
For that briefing note falsely asserts, in line with a number of audit guides produced since, that the discounts in question are being called 'questionable' on the basis that the electoral register suggests that there might be more than one resident. But it is perfectly legal to be in receipt of an SPD when more than one adult is resident.
How can such an inaccurate use of electoral register information be done 'in accordance with the law'. The whole argument is not 'in accordance with the law' ie council tax law: it was at odds with it.
But the briefing note says
We have so far restricted this initiative to unambiguous matching i.e. where the SPD applicant has declared there to be nobody over 18 in the household.
This is not true. The NFI has been told a number of times by lawyers in different contexts that this is not true, but appears not to have taken the message on board. Whether this is deliberate or not is an open question.
Given how many of these additional voters were disregarded, and some will have had their sole or main residence elsewhere, the bare figures in terms of questionable discounts are highly misleading. Such evidence as is available suggests that the majority of people highlighted as 'questionable' using this method are in fact fully entitled, and are not making any false or misleading claim. The misleading claims here emanate from the NFI.
It is incorrect to assert that the pilot schemes showed that anything like this was being lost nationally. If Mr Lewis was told that this was the case, then whoever instructed him was not doing a very good job of it. Remember that the so-called 'evidence' relied on in coming to these figures is the mere presence of a second adult of the electoral register. This does not indicate lack of entitlement, or any maladministration by the council, or a failure to declare changes in circumstances which the law requires the taxpayer to tell the council tax department about, and so it is bad statistics as well as bad reasoning and bad law to jump to the conclusion that numbers of 'hits' equate in any way to lost revenue, leave alone lost revenue.
Given that NFI documents dating from before the date of Clive Lewis's report contain legally inaccurate assertions about how council tax discounts work, there is a clear prima facie case that he was instructed on the basis of incorrect beliefs about this whole matter
The Optimal Time
I am also instructed that the optimum time for comparison of data from the full electoral register and the council tax records is when the electoral register is re-published in December each year.
Thus, for this financial year, the optimum time for comparison is December 2007.
If you want lots of abortive investigations into innocent law abiding families, yes, this is the optimum time. This is clear evidence that Mr Lewis's instructions were problematic, and it strongly suggests that he did not bother to check the law underpinning them. Or else he is simply very relaxed about thousands of entitled people being labelled as potential fraud cases when there is not one shred of evidence. But on the evidence of his own opinion, the former appears to be the correct explanation.
Possible Cases of Fraud
Paragraph 2.3(e) and (f) provide that “wherever practicable”,
information that data may be disclosed for auditing purposes in order to identify
possible cases of fraud should be given to the data subject prior to the initial
collection of the data.
Since the NFI denies that this exercise labels people as being in inconsistent situations, leave alone as potential frauds, this comment is in law irrelevant. But since Mr Lewis says it, I think it is possibly true.
Odd that since having a legal briefing on council tax the NFI has taken to denying that the hit lists are lists of 'possible cases of fraud', and even to denying that there is any requirement that all cases should be possible cases of fraud (except if they drafted the explanatory notes to the draft local audit bill, in which case they need their knuckles rapping for misleading Parliament). So here we have in effect evidence from a senior lawyer that the Audit Commission has been pack pedaling and contradicting itself.
The Single Occupant Discount which does not exist.
The provision of the electoral roll and council tax records for a particular local
authority will enable the appointed auditor for that authority to match or compare
the two sources of data. The electoral register may indicate, for example, that
there is more than one person living in a particular property but the council tax
records may show that the person liable for paying council tax is claiming the
single occupant discount.
Here we run into the problem that the NFI data matching expert has asserted in writing that when Clive wrote 'but' here what he said/meant was 'and'.
For it is perfectly legal to be in receipt of a discount when more than one adult is resident. So we are left unclear about what Mr Lewis meant by this paragraph as the same expert has said that he chose his words very carefully. Does or does not a senior barrister grasp the plain difference between the words 'and' and 'but'? If not, then perhaps the Commission should have consulted somebody else.
The best way to look at the problems presented by this is to ask what a sensible barrister with some sense of social conscience might have done if presented with the briefing notes and the audit guide and asked to advise on their legal accuracy and the consequences for the NFI of them being wrong. I suppose that if asked for legal advice on a personal basis by a responsible person, he might have told them simply to shut up and hope nobody asked them any awkward questions.
On this basis, I believe that Clive Lewis was not aware that his opinion would be used as the basis for arguments by Audit Scotland and the NFI to the effect that there is something called a sole occupant discount and you can tell by looking at the electoral register that the accounts are wrong, which is the central thrust of his whole advice.
But he must know by now, and one could consider he has a duty since the opinion has been put on the web site of Parliament and used as a campaigning tool by the NFI, to put the matter right.
Unless Clive Lewis believes that it is lawful for auditors to carry out statistically based fishing trips in search of fraud which is what this is: it is one fishing trip inside another, or as somebody from CIPFA put it recently, a way of limiting the sample size in one's pro-active search for fraud.
But I doubt this because in the same opinion Clive Lewis sets out the powers and duties of auditors, including those auditing the accounts of councils. And this says exactly the opposite. In fact his whole argument only makes sense if you take it that he falsely believes that there is something called a 'sole occupant discount' which only those living literally alone can claim and that the electoral register as near as dammit shows lack of entitlement.
If you read his opinion in the light of the facts of council tax discount law, what he is saying is that auditors may not have access to the full electoral register. Because they cannot tell from it that a council is failing in a duty to obey the law on the financial side of things, or that revenue has been foregone.
I think this opinion may explain the tendency evident on council web sites to describe disregard categories as 'discounts' when, clearly, they are disregard categories existing within the provisions setting out one discount. For it mentions 'discounts' and refers to the sole occupant discount as if it were one of these.
This may explain the fact that Scotland, where Audit Scotland used Clive Lewis's opinion as a lobbying tool, is even fuller of councils falsely asserting that you cannot be entitled to a single person discount when more than one adult is resident. Edinburgh is a good example and Experian is involved in this one.
