Select Committee on Treasury Minutes of Evidence

Examination of Witnesses (Questions 200-219)


26 OCTOBER 2005

  Q200  Mr Todd: That does occur.

  Dawn Primarolo: Indeed.

  Q201  Mr Todd: To wrap this up, you refer to the learning experience of one of your team, if you like, gathering that together. I think for everyone's benefit it would be useful if such a document, if it was prepared in that form, was publicly available so we could all learn from that. You made a perfectly fair pointed reference to the fact that Parliament did not scrutinise these aspects when it considered the introduction of tax credits. Perhaps it should have done and perhaps it would do in future if it had a greater degree of shared experience about what had gone wrong in an experience of this kind. It would be very helpful if that was the case.

  Dawn Primarolo: Indeed, just to clarify—I think you were touching on the question of the resilience of the system.

  Q202  Mr Todd: Yes.

  Dawn Primarolo: That will have to be when everything is settled with EDS that I can raise that and ensure the Committee gets what it is asking for.

  Q203  Ms Keeble: I want to ask some questions about error and fraud. The Comptroller and Auditor General recently qualified his audited opinion on the 2004-05 accounts because of the probable unacceptably high level of claimant fraud and error which is likely to have been about 3.4%. In your opening statements you referred to three different ways in which you are looking to improve the position about improved communication, reducing error and improved procedures for recovery. You have said quite a lot about communication and recovery but how about changing the procedures so you can reduce the errors, can you say any more about what you are doing there?

  Dawn Primarolo: I can say what the procedures are that the Department undertakes now and where we are with regard to that assessment and, of course, the Comptroller and Auditor General's view qualifying the accounts of the Department. This was a matter for quite a lot of discussion when the Bill was going through and, indeed, I think we have had discussion here. Firstly, the Department has to verify and check all claims before payments. Secondly, it has to do a risk assessment on the claim and report changes with examination of cases where risks are identified. Thirdly, it will disrupt or terminate where fraud is suspected, and there is always some crossover there, and ensure that financial penalties and prosecutions from these serious cases are in place. Now, on what the Department has done, and how we got to the 3.4%, and that is an early estimate going through an analysed representative sample, which is the way it is done on the finalised claims, it came in at 3.4%-ish, let us finish the work. Clearly the NAO qualified the accounts because the NAO benchmark is 1%. Now the levels are lower than they were in their predecessor benefits it would appear but obviously at 1% it has to be a considerable amount of pressure to get down to that, although DWP accounts have been qualified since 1986. What the Department has to do, we have to sort out as part of that what is going on now with regard to disputed overpayments because we have to try and separate what is a real disputed overpayment and whether there is something else going on, to ensure that all those checks and verifications are working and the difference between the estimated and 1% benchmark, we will have to review whether that is severe enough to drive down the percentage even further to 1%.

  Q204  Ms Keeble: In the sample you are talking about, from the information we have got, that work is not due to be completed, in fact, until the spring of next year and then do you think it is irresponsible to have such a huge programme—£30 billion programme—rolling out without having some reliable data collected quite quickly on the local area of fraud, given that it will not be completed until next spring?

  Dawn Primarolo: Yes, but the data has to be collected from the finalised awards of 2003-04. It has to be scrutinised and tested and then it has to go through the requirements to ensure the data is correct. That data simply was not available until those awards were finalised. The approach, and there was a great deal of discussion about this, is to build automatically into the system pressures which would act on a downward pressure on that fraud. What we need to do from those figures is then ascertain exactly what percentage is error, and therefore cannot be collected, and what is fraud which we might need different procedures for.

  Q205  Ms Keeble: I want to ask something about the fraud but, first of all, presumably there is some sampling mechanism built in right from the outset, from the design of the system?

  Dawn Primarolo: Yes.

  Q206  Ms Keeble: Why did it have to wait until the 2003-04 figures were completed, given that those have taken a long time to complete and some are still disputed?

  Dawn Primarolo: Because tax credits are determined on P60s from the previous year and then are projected forward to say if your income stays the same or only varies by a maximum of £2,500 and there are no changes in circumstances, that is a provisional award to you, and then it is reconciled at the end of the year and finalised, which is how much did you actually earn, and therefore you have a statement of how much exactly individuals were entitled to. So while the process was going on—and I am happy to send a more detailed paper on this—those steps are in when people apply for the tax credits to try and act as a downward pressure and to eliminate those claims that are considered fraudulent or without substance.

  Q207  Ms Keeble: What I am concerned about is that one of the issues about tax credits is that they are very generous, which is great for people on low incomes, but it obviously means that the potential for fraud is quite substantial. What you virtually end up saying is that for the first two years it is very hard to get the checks in place because of the way the system operates.

  Dawn Primarolo: No, I am not saying that. Firstly, the vast majority of taxpayers are perfectly straightforward and honest. Let's be correct here and have that on the table. Secondly, what I am saying to you is I am recognising that the system as it operates, automatically, through verification, risk assessment, disruptional termination and finally penalties and prosecution, if necessary, has a way of stopping that fraudulent activity succeeding, and then the finalised awards, which are looked at in detail in the sample is a sort of double check on that, and then it gives a percentage. Quite rightly, the NAO have said the benchmark is 1% and that is what you have to get to. To make it sound as if the system is going ahead and nothing is happening in between times is not the case. In fact, sometimes we cut people's money off and they do not like it and then they go to their MP and say we have cut their money off.

  Q208  Ms Keeble: I understand that. I want to ask a question specifically about fraud. I completely accept that for most people they are going to put in completely straightforward claims and for most people the payments will be accurate, they will be very welcome, they will be generous, they are particularly important for women (although there has been the problem about overpayment) however, what checks are actually done particularly on identity fraud? For example, in particular where people might have national insurance cards or whatever which are not valid?

  Dawn Primarolo: That would be within the verification checks of all claims before payment, and if you would like I could ensure that you had a paper on that. We have got a system which is recognising a number of things which also need to be verified. 700,000 children are born a year and that could lead to an increase. Half a million young people get to the age of 18 and move into employment and that changes it. There are 120,000 lone parents moving into employment—good—27,000 of them from part time to full time. So there are lots of good things going on in the system and we have to balance what is good. It is the time old debate that is had about a benefits system per se in making sure that it has the right checks in place to deal with those who may decide they want to commit fraudulent activity, whilst not making it more difficult for those who are straightforward and claiming what they are entitled to. On verification, that is done, that is why we ask people to check forms. Alright, sometimes the form says they have got more children than they have or they have not got as many children as they have and that is where the verification and the links between the claimant and the Department come in.[3]

  Q209 Ms Keeble: And national insurance cards in particular?

  Dawn Primarolo: Well, it is done on a national insurance number and we have to have the national insurance number as part of the claim. Some people claim and the forms are not complete. They are not processed until they are which results for some people in delays to their payment.

  Q210  Ms Keeble: There is a problem about those cards being forged, is there not?

  Dawn Primarolo: You are asking me a wider question about the security of the national insurance system. I am telling you that the verification checks are in place. The problem in asking the Department to give you information is that to reveal exactly how the Department does its risk analysis to combat fraud is as good as saying "This is how you might be able to get around it." I can ensure that the Department supplies more information than I am equipped to give you this afternoon but that will be limited also and I am sure they will put the caveat here so as not to say "Here is a roadmap on how to get round that."

  Ms Keeble: I understand that.

  Chairman: Lorely Burt?

  Q211  Lorely Burt: I would like to ask you one or two questions about the Ombudsman's report please. The first one is can you tell me why the Department refused to accept the Ombudsman's tenth recommendation that the MHRC consider writing off all excess and overpayments caused by official error which occurred during 2003-04 and 2004-05?

  Dawn Primarolo: Because the Department is under a duty of care to all taxpayers as well as those receiving overpayments of tax credits, should that happen. There are many people, many homes, who pay back their tax credits, who recognise, despite the fact that there might have been an error on behalf of the Department, that they should not have had the money and they are prepared to pay it back. It was simply—and I said it in Parliament—balancing the duty of care to all taxpayers and the duty of care to individual tax credits claimants, which is a difficult balance and that is why we use the test we do. It is exactly the same one as in the tax system. That was the response to the Ombudsman. I have not heard any other response coming back from her yet but discussions between her and the Department obviously are going to have to be on-going and I am very much looking forward to her being involved with the Department, as she is, in advising on and addressing the issues that she says we need to take forward.

  Q212  Lorely Burt: I am glad to hear that. I wonder if I could turn to the eleventh recommendation that you consider adopting a statutory test for recovery of excess payments and overpayment of tax credits, consistent with the test that is currently applied to social security benefits, with a right appeal to an independent tribunal. What is your position on that?

  Dawn Primarolo: The Ombudsman asked us to consider that within the tax credits system and therefore that must be taken forward within consideration of what is called COP26. I stress again that the tax system is not the benefits system. The tax system is the tax system and it works on an automated process for very good reasons, particularly for in-year recovery, so what I expect—and I know that the Ombudsman would want to be, rightly, involved in this—is for that matter to be addressed and the Department to respond accordingly, and that is what they are doing, and they will tell me as well.

  Q213  Lorely Burt: From what you are saying it sounds like we should not be too hopeful.

  Dawn Primarolo: Frankly, I am convinced that in a system that is so large we have to have automated recovery, and that was very much part of the discussion. If somebody's income changes dramatically in year, the best way to stop any overpayment is to address their tax credits, how much they are entitled to for the whole year given the change in their incomes, and in that way you do not get an overpayment, so this has to be seen as part of what the principles of the tax system are operating. It is not for me to say whether I am hopeful or not. It is for me to ensure that these issues are properly addressed within the broad prospect of the policy and whether or not that can be taken forward. I am not speculating.

  Q214  Lorely Burt: I appreciate all that you say, however, there have been so many problems for so many thousands of people, and I think this was why recommendation ten came in because of the number of people who had, in all good faith, given information to the Department, the Department have made the errors, the money has been given to these people, they have supplied all the right information. It is not their fault, and they have spent it and they do not have the money then to pay it back to you. I think this is the reason why these two recommendations, 10 and 11, were made.

  Dawn Primarolo: I agree and all I can say to you is can we keep it in perspective, please? There are 6.1 million families and, yes, you are absolutely right that there are a number of families for which this has not worked as it should and we have to address that. You are talking about errors, failure for administrative reasons or computer reasons, and that should of course be put right. Then there is a separate issue of if the system is working as it should, is there a need for an appeals procedure? If there is, what would that look like? That must be taken forward as part of that consideration. That is all I am saying to you. I get a little frustrated when, rightly so, the terrible service that some have received—and it is a small percentage in the tax credits; 6.1 million is a large number of people—is presented as the reality for all tax credits recipients, and it is not.

  Q215  Lorely Burt: Is it not the case, though, that over 30% of people were overpaid in the first year of the scheme?

  Dawn Primarolo: Yes and £1 billion of the £2 billion was accounted for by households who had earned £10,000 more a year. That is a pretty significant change in circumstance. The issue is did the Department know and not act on it in a timely way or did the Department not provide enough support and information to the tax credits claimant that they should have notified at an earlier stage. Cut the other way, half of the 1.9 million who were overpaid were overpaid by £570 something or less in the entire year. Then alongside that—and I think Mr McFall had a case—there are also some people who, for various reasons that are inexplicable, received considerable amounts of money.

  Q216  Lorely Burt: The Ombudsman said that there was systematic maladministration over the automatic recovery of overpayments. When we met David Varney he flatly contradicted her and said that there was not. Is it for him to say or for the Ombudsman? Can you give a yes or no answer to the question: is there systematic maladministration?

  Dawn Primarolo: It is for the Ombudsman to say whatever she decides she wants to say. I would quite like to know, and will be enquiring exactly what was meant by "systematic" because that only appeared as a suggestion last week in committee, and that is something that I would be very concerned about as to how she came to those conclusions and what her test is for coming to that conclusion, and then I will be able to form a judgment. I do not have that information currently and I had no idea before it was actually said in committee and she did not say that in her report. Indeed, she said herself in her report that she does not consider the system to be incalculable for everyone, or words to that effect, and that for a significant number it is working well. She rightly concerns herself, as we all do, with those where it does not work well. I cannot comment on a statement that was made when I do not know on what basis it was made and what are the criteria for making such a suggestion, and that it is something that was not said in her report.

  Q217  Lorely Burt: Just one final question, if I may. Do you think it is simply a matter of tweaking the administration and improving the IT and then everything will be fine, or do you think that the client group that it is chiefly intended for, which is those on very tight budgets and low incomes, the uncertainty and the inherent overpayment is always going to be problematic?

  Dawn Primarolo: It is not just intended for, as you called it, and I did not quite catch the last word, the very vulnerable. Nine out of ten families benefit. It is hugely superior to the fixed, time-limited snapshot award that was its predecessor, which was also widely criticised, and family credit before that. It has specific objectives in terms of removing stigma, moving people into work, eradicating child poverty and making a contribution to that objective. Yes, I think this system is the system to do that. The challenge now in the immediacy is to get the errors out of the system, get the IT working properly and to make sure that people are getting the service that they are entitled to—the right money at the right time—and proper communications from the Department, and of course deal with the issues, should they arise, of error and fraud. Those are my objectives and I will work with whoever is prepared to work with me on those objectives in order to improve the service.

  Q218  Mr Love: Can I endorse that last statement you just made about the system and, indeed, can I congratulate the Department on reaching 6.1 million families because I think that needs to be said as well. I wanted to talk a little bit about the design of the system because there has been research carried out by One Parent Families which shows that the level of understanding of this system is quite high, but picking up the point Jim Cousins made earlier on, they found that 47% of the people they polled had somewhere between two and seven changes in their circumstances every year, 36% had at least one change, and only 17% had no changes at all. Recognising that the current design of the system under which we are operating appears to be impacting most on vulnerable families, is that a particular concern for the Department at the present time and are you looking at that?

  Dawn Primarolo: Firstly, the report that has recently been published by One Parent Families, which I have recently received (and I do want to look at it in some detail and have discussions with them and I have not been able to do that yet) I would say, yes, one of the issues is this question of change of circumstance, how frequently it occurs. People moving into work—thankfully and because of the brilliant stewardship of the economy by the Chancellor—is much, much higher and the turnover of those in work is higher than we had anticipated. That is a challenge. It is a challenge in terms of looking forward so, yes, the consideration—and I would like to speak to them in more detail about the size of their sample and looking at the points they are making, for us to understand exactly, because whilst informing applicants for tax credits of what would be a change of circumstances it does also mean that we have to be advising them clearly and understand what is going on as a Department, and we need to take that forward. I cannot comment at the minute but I think it does raise some important points, which have been raised by other organisations.

  Q219  Mr Love: The Department itself has not carried out any research in this area? It seems an obvious thing to do.

  Dawn Primarolo: We are trying to look at that now but at the same time our priority is to sort out what we have been asked to sort out already so, yes, an analysis of what might be causing this churn in change of circumstances is something that we do need to look at, and I have asked that it be considered.

3   Ev 44 Back

previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2005
Prepared 15 December 2005