Official Report (Hansard)

Session: 2011/2012

Date: 19 January 2012

PDF version of this report (226.78 kb)

Committee for Social Development

 

Welfare Reform: Fraud and Error

 

The Chairperson:

I welcome John O’Neill and Conrad McConnell.  Obviously, we are here to deal with the departmental briefing on welfare fraud and error.  I remind members that the relevant papers are in their folders.  It is over to you, gentlemen, for the presentation.

 

Mr John O’Neill (Department for Social Development):

Thank you, Chairman.

 

The Chairperson:

The subject had to be deferred at the previous meeting, so we did not discuss it, but may we get clarification of why we deal with fraud and error together?  They are two different things, so give us an understanding of why we deal with both on the same basis.

 

Mr Conrad McConnell (Department for Social Development):

We tend to group them together, Chair, because it is all about loss to us.  It is the wrong benefit, whether it is too much or not enough.  When we come before the Public Accounts Committee (PAC) and other Committees, they tend to —

 

The Chairperson:

I am making the point that fraud is deliberate.

 

Mr McConnell:

Absolutely.

 

The Chairperson:

Error is not.

 

Mr McConnell:

Yes.  We always make that distinction between the two, but we tend to group them together for the purpose of briefings and talking about a loss or underpayments or some combination of the two.

 

The Chairperson:

OK.

 

Mr J O’Neill:

Again, thank you, Chair and ladies and gentlemen of the Committee.  It is good to have the opportunity to talk to you.  This is part of a series of briefings around welfare reform.  I understand that this briefing was originally scheduled to be held before Christmas.  You had a briefing note from us.  Today’s presentation will expand on that and set the context.  I am happy to run through that for you.  Does everyone have a copy of the presentation?  Unfortunately, it is in PowerPoint.

 

The Chairperson:

Does everyone have that?  Yes.

 

Mr J O’Neill:

The GB strategy was developed jointly between the Department for Work and Pensions (DWP) and Revenue and Customs in late 2010.  The strategy’s scope covered social security benefits, housing benefit and tax credit.  The aim of the strategy was to put forward measures and proposals around further reducing the level of public funds lost through fraud and error. 

 

To show the Committee the scale of what we are talking about:  the strategy confirmed that within the £190 billion spend on benefits in GB there were losses of more than £5 billion a year.  You will see from the presentation that that equates to 3% of total expenditure.  To break that down a bit further: that equates to just over 2% of Department for Work and Pensions spend and just under 9% of Revenue and Customs spend on tax credit. 

 

The corresponding percentage position in Northern Ireland, including housing benefit, is 1·3% total losses against expenditure.  Losses of public money are well down.  In 2004-05, the total loss was £120 million a year, and, in 2010, it was just under £67 million.  Obviously, that still represents a significant sum of money, and the Public Accounts Committee has looked at that issue on a number of occasions.  You will see that that Committee recognises the progress that the agency is making in tackling losses but requires more to be done.  Specifically, it expects tackling benefit fraud and error to remain a key priority for the agency.  Agency losses are down further at just over 1%, but we are not complacent because that still represents £48 million of public money.  We need to continue to maintain the momentum and to keep tackling fraud and error.

 

I turn to the area of fraud.  Hopefully, Chair, this will address the question that you raised initially about why we consider fraud and error together.  We put in place measures to tackle customer fraud and customer error.  We know that that can be a sensitive subject, so, at a high level, we have set out some numbers from 2010 on our approach to tackling fraud and error.  Those numbers need to be set in the context of the fact that the agency deals with a total caseload of over 600,000 customers and makes over 800,000 benefit payments each week.  We consider that the agency tackles this in a very balanced way.  We target around 14,000 case reviews every year through case interventions, and you will see that, in 2010, we made just over 800 changes.  On the issue of compliance, we have more specific information that indicates that there is a risk of a change of circumstances.  It challenges the customer’s information during a face-to-face interview.  We project that, in 2011-12, we will carry out 3,000 compliance interviews, and, from that, we expect to make around 1,000 changes.

 

Fraud investigations are reserved for more serious cases, where there is a higher risk of intent or deliberate behaviour.  We carried out just over 4,000 fraud investigations in 2010-11, and that resulted in 1,136 changes.  Of those, 542 went to court.  The key message about tackling customer fraud and error that I want to leave you with is that fraud investigation itself is reserved for deliberate frauds.  Of those that we deal with, around half are dealt with outside the court process.

 

I thought that it would be helpful to put prosecutions into the context of the level of sentencing involved in the cases that went to court.  There were 542 cases that went forward and resulted in conviction, but that needs to be set against a caseload of over 650,000 customers.  Almost half of the 542 cases involved fraud of between £5,000 and £30,000.  In fact, the highest level of fraud last year was just under £64,000.  It reinforces the message that fraud represents serious, deliberate criminality involving large sums of money.  The aim of the Welfare Reform Bill provisions is to help to tackle such criminality, and there is a strong focus on prevention through deterrents.

 

As for official error, we obviously recognise that loss also occurs through staff error, and, as with the customer fraud and error figures, losses from staff error have reduced steadily.  They are down from around £34 million in 2004-05 to just over £20 million last year.  Staff performance is subject to continuous close scrutiny, and there is a very strong emphasis on accuracy.  All the main branches in the agency have accuracy improvement plans to drive performance.  The financial accuracy figures are reviewed every month by the agency management board.  There are around 76,000 targeted checks on top of the normal management checks that are done and the controls in the day-to-day running of the business.  To give you a comparison with the customer fraud figures: we check, in a rolling year, about 15% of our caseload in order to make sure that cases are accurate.  The overall level of financial accuracy in the benefits system is 99·2%.

 

We have an independent standards committee that looks at the standards of decision-making in the agency.  It is chaired by Professor Eileen Evason, who comments annually on the performance of staff in terms of their accuracy and decision-making.  Her report for last year commends the strong performance, in the context, obviously, of an extremely complex benefits system and IT systems, which, I understand, members of the Committee have seen for themselves.  Official error has improved, but we are not complacent, and we remain focused on staff accuracy as well as customer error and fraud.

 

That sets the context for the proposals in the Welfare Reform Bill.  I will hand over to my colleague Conrad McConnell, who has responsibility for the implementation of the fraud and error strategy in the agency.  He will walk you through the key proposals in the Bill.

 

The Chairperson:

OK, John, thank you.

 

Mr McConnell:

I will explain what the strategy is about.  The GB strategy was published in October 2010 and sets out what DWP and HM Revenue and Customs (HMRC) want to do to build on tackling their problem.  We want to mirror that in order to bring our figures down as far as possible.  The thrust of the strategy is about trying to bring together separate organisations that are tackling fraud and error individually and to make the approach much more joined-up and focused.  The strategy includes elements such as a single fraud investigation service, which would take away the existing situation in which the housing authorities investigate housing fraud separately, DWP investigates benefit fraud separately and the HMRC does its own fraud investigations.  The idea is to bring all that together into a single fraud investigation service, in which one investigator can go to a home, a household or an individual and investigate any potential fraud for all aspects of income, rather than individual investigations being done by different people in different organisations.

 

The key to tackling fraud and error is to target our activities as well as we can.  One way to help that is to better inform ourselves about risk and to target that risk.  If we invest money in investigations or other activity, we can target areas that are more likely to contain fraud or error.  Therefore, the proposal is to make use of what is called a centralised risk and intelligence service, which will, hopefully, bring together better data matching and information to help us to detect the cases that have fraud or error in them.

 

The strategy proposes to introduce slightly firmer penalties for benefit fraud.  Again, I want to make a distinction between fraud and error.  This is about fraud.  We have penalties for intentional behaviour, and we want to toughen up some of those.  I will come to those in a moment.  Another aspect of the strategy is to increase the amount of money that we can get back where fraud and error occurs and to reduce the overall debt stock in government. 

 

It is not all about fraud, of course; it is also about error.  There are a number of proposals in the strategy about better tackling error and cleansing some of the live load.  Some benefits have higher instances of error than others.  The strategy, in joining up all those roles, will help us to prepare for universal credit, when the benefits systems will come closer together.  We will have a better joined-up fraud and error system to underpin the administration of the new joined-up universal credit benefit system. 

 

I will break things down in a little more detail.  I mentioned the single fraud investigation service, which will bring together individual investigation authorities under one authority.  On the penalties side, there are a number of proposals in the Welfare Reform Bill.  For example, it will remove formal cautions, which will result in a system that imposes administrative penalties — what might be called internal fines — or a court process as an alternative.

 

At the minute, we do not have a minimum for the financial penalty that we impose, so the proposal is to introduce a minimum of £350.  Under the loss of benefit regulations, which is a separate agenda item this morning, it is proposed in the strategy to increase some of those penalties.  At present, the regulations introduce a four-week loss.  The strategy proposes to increase four weeks to 13 weeks and to introduce second and third strike penalties up to a maximum of three years for really serious frauds.  The strategy also proposes the introduction of a civil penalty in the case of people who have negligently not reported changes in circumstances that may not extend as far as behaviour that constitutes fraudulent or intentional criminality.

 

The presentation shows that there are issues around debt recovery.  It is proposed to increase the maximum rate at which we can recover overpayments and to consider whether we can recover money overpaid by staff.  The Committee may have views around some of that.  Another point is fast-track recovery, for example, directly from employers or by using the pay as you earn (PAYE) system.

 

I want to make the point that the penalties and sanctions and so on are primarily about putting people off committing fraud.  Our aim is to put people off committing fraud rather than to impose any of the penalties and sanctions.  I just make that general point.

 

Other changes that are planned as part of the strategy will not necessarily be part of the Bill because they are not in the formal legislation.  However, I will make the Committee aware of some of those changes as part of the overall picture.  I mentioned the central integrated risk intelligence service (IRIS), which, again, is about helping us to identify risk and to target our activity in the right places.  GB also intends to do a little more around communication to get the message out about fraud, the problems it creates and what it does to the overall benefits system.  Ultimately, the system is designed to help people, and we want to try to maintain the integrity of a system that helps people, rather than have it subject to fraudulent activity.  There are telephone numbers across the operation for GB, HMRC and DWP, and the aim is to make it simpler to report fraud through the use of one phone number.

 

There are another couple of points to make at the end of this presentation.  Case cleansing is about error — if there is error in the system, can we cleanse some of the live loads and get some of that out?  John mentioned the levels of fraud and error.  As we move towards universal credit, we want to make sure that we have one system that measures the overall levels of fraud and error, whether that relates to HMRC, DWP or housing benefit, in a single way so that we can properly benchmark them against each other.  Finally, the ultimate aim of all this is to pull together all our activities to prepare for universal credit, under which it will become a single administration service anyway. 

 

So, those are some of the details.  At this stage, they are still quite high level.  Obviously, more detail will emerge when the Bill comes through and the regulations thereafter.  However, our aim is to try to give you a flavour of the thrust and intention of some of the changes that are planned.

 

The Chairperson:

OK, Conrad, thanks to you and John for that.

 

Mr Brady:

Thanks very much for the presentation.  I must say that it sounded familiar.  On the key points in your briefing paper: is there a breakdown of how much of the £66·1 million in fraud and error in the North is fraud and how much is error?

 

Mr McConnell:

Yes.  The figures that I have purely relate to the Social Security Agency (SSA), which leaves out housing, but I will give you those because it is a similar sort of breakdown and will give you a flavour of things.

 

Mr Brady:

I ask because the emphasis is always on fraud and people being targeted.  All the campaigns, which cost a lot of money, highlight alleged or actual fraud.  However, little emphasis is put on departmental error.  When I worked in the Department many years ago, if an official error was made, it was raised as an overpayment against the member of staff who had made it.  It did not actually affect them, other than possibly affecting their promotion and stuff like that because it was on their file.  Obviously, somebody who made a number of mistakes would eventually be penalised.  They did not have to pay the money back, but, as far as I am aware, that has all gone. 

 

I have another point to make before you give us the figures.  Debt recovery is to include official error repayment.  In other words, if an individual gives the Department all the information that it requires and somebody makes a mistake, the individual will be penalised because somebody has misinterpreted something or failed to properly do what they were supposed to do.  We talk about IT systems, but they are only as good as the information fed into them. 

 

As you know, there has always been a bone of contention about failure to disclose or misrepresentation.  From my experience over the years, I know that there were many cases where people could not disclose something that they did not know.  Misrepresentation is a different thing, because you were being accused of wilfully giving wrong detail or facts.  It is interesting, because there seems to be an emphasis now on what will happen if an official mistake is made.  In such cases, the claimant is going to pay for it, even though it is not their mistake.  I am not sure what the rationale behind that is.  It is obviously to recover money, but it is unfair.

 

Mr McConnell:

I will take your first question first and give a breakdown of the source of the problem.  Let us take SSA benefits and the loss of SSA benefits.  Last year, we lost 1·1% of our overall spend.  That is made up of 0·5% fraud, which is just less than half of the overall loss; 0·4% due to staff error; and 0·2% due to customer error.  That gives you a flavour of the breakdown of where the losses are coming from.

 

Mr Brady:

Thanks for that.  The reason I ask is that the Committee got figures in 2008, and the actual official error rate was more than the alleged fraud from claimants.  Again, I go back to my point: a huge emphasis is placed on people who are defrauding the system when, in fact, the official error is as much as, if not more than, it was before, and it continues.  It is something that needs to be redressed.

 

Mr McConnell:

If we include underpayments that are due to official error and look at the overall financial accuracy, we see that there is a rate of 0·8% of underpayment and overpayment, which equates to 99·2% accuracy in a system that has been commented on many times here, in DWP and by the standards committee and others, due to its complexity and how difficult it is for staff to administer.

 

Mr Brady:

I understand that.  You talk about the percentage of underpayments, but those are the underpayments that are detected.  There may be thousands of other people out there.  You talk about a caseload of 600,000.  There may well be a lot of people within that caseload who are being underpaid, and that is not contained in the figures because you do not know about it.  I just wanted to make the general point that it is not all down to fraud; it is as much about official error.

 

Mr McConnell:

I hope that the figures that were presented earlier set out some of the scale of what we do to tackle the overall problem.  We have thousands and thousands of checks each year to keep that 99·2% accuracy as high as it can be in a system that is complex.  However, at the same time, we cannot disregard other behaviour.  We tackle staff error and customer error in a way that does not involve courts and all of that, but we have the court processes and investigations for the serious fraud cases.

 

Mr Brady:

Nobody is condoning social security fraud.  That is not the issue.  The issue is how it is dealt with fairly.  The figures that you have given us for prosecutions, etc, are relatively small compared with the number of cases.  In my experience as someone who worked in the system and outside it for many years, a lot of the information that the Department gets is wilful information from people who have wilful intent, and that immediately triggers an investigation.  There may be fraud in many of those cases, and, in many, there may not be.  It is an inaccurate gauge as to what is happening.

 

Mr McConnell:

I should clarify that the fraud figures and the error figures are not based on our activity for tackling fraud.  They are based on an independent measurement system, in which we sample the live load and test cases for instances of fraud or error.  What the fraud investigators do and find is not the measure of the level in the system; the measure is a random sampling process.

 

Mr Brady:

But it is an inclusive system, obviously.  If you are saying that a particular amount is lost to fraud, there has to be some basis for that figure.

 

Mr McConnell:

Yes, there is.

 

Mr Brady:

I presume that it is not an arbitrary figure.

 

Mr McConnell:

No.  Cases are tested in a random sample from a pool of cases, and the figures produced are statistically valid.  Cases are then checked for fraud or error. 

 

Mr Brady:

It is just that I find that the situation from 2008 until now does not appear to have changed that much.  Official error is still as high as alleged or actual fraud.

 

Mr McConnell:

The overall level of loss is down, however.

 

Mr Brady:

That is good, but —

 

Mr McConnell:

It is down from 1·7% to 1·1%, and the purpose of all this is to try to keep it down or to improve on it. 

 

Mr J O’Neill:

Let me add to what Conrad has said.  The key point that we are trying to make is that we take a balanced approach to this.  We do not focus solely on customer fraud and error; we focus on official error, and we have a checking regime to try to improve the accuracy of our work.  I would love to tell you that the system is perfect, but you recognise, as the previous Committee recognised, the complexity of the system and of the IT system that supports it.  We will continue to strive to improve its accuracy. 

 

Mr Brady:

With respect, you talk about a balanced view, but, in fairness, it is a very subjective view as opposed to an objective one.  That is my experience of it, which goes back a number of years. 

 

The Chairperson:

A question was asked about your focus on reclaiming money from people who have been in receipt of overpayments.  That was not answered.

 

Mr McConnell:

The strategy is about trying to deter fraudulent behaviour, and one of the elements of that is to try to recover money.  The strategy sets out the principle, and that can be debated through the Bill, the regulations and everything thereafter.  However, the general principle of the strategy is that, aside from the issue of fault, is it ultimately right for a person to receive money that they should not have received?  The rules should not have allowed it to happen, but it happened because someone made an error.  You talked about the member of staff making an error.  The question is, therefore, whether it is right to allow the person to hold onto that money, even though the rules say that they should not have received it.

 

Mr Brady:

Let me make one point.  Sometimes, the bigger picture is overlooked.  In the past, I have represented people in cases where an overpayment of £150 was raised against them.  Those people said that it was not their fault.  The cases went to tribunal, which probably cost £1,000 or £1,500.  That money may have been from a different budget, but it all comes out of the public purse.  It does not make sense. 

 

Mr McConnell:

As I said, the proposals can be debated.  

 

Mr Copeland:

It is always difficult to follow Mickey Brady because he tends to ask all the questions that you have thought of, so I am struggling slightly.

 

I want to check a couple of things.  I understand fraud, and I understand error.  However, I think that error — Mickey has alluded to this — should be split into departmental error and customer or client error.  Can you give me a breakdown of what the numbers and percentages are? 

 

Mr McConnell:

It is.  Yes.

 

Mr Copeland:

Can we then look at the figures on which we are basing Northern Ireland relative to the rest of the United Kingdom?  Does the figure for GB include housing benefit?  It does in Northern Ireland, because it is paid through the system.  Are we comparing like with like?

 

Mr McConnell:

The figure for Northern Ireland is a 1·3% loss, including housing benefit.

 

Mr Copeland:

Does the figure for GB include Northern Ireland?

 

Mr McConnell:

No.  The GB figure is a DWP figure.

 

Mr Copeland:

Right.  Does it include housing benefit?

 

Mr McConnell:

It does.

 

Mr Copeland:

So, we are comparing like with like. 

 

On the matter of encouraging people to give information about suspected fraud, do you take into account frivolousness and vexatiousness?  There is a postcode in east Belfast that identifies 26 houses, and, at some stage, the vast majority of those 26 addresses have been the subject of actual and erroneous complaints about benefit fraud.  In some cases, that has caused severe distress to individuals, most of whom — to be fair — are employed.  Do you take steps to identify the motivation of people who make frivolous and vexatious complaints?

 

Mr McConnell:

We have a system whereby we try to sift out from all the referrals that we get.  We would not want to investigate every case, because it would not be good value for money for us to do so.  We try to sift out the cases where it looks, on the surface, as though the evidence is strong and the information is sound.  We would not want to annoy people to whom we have no need to speak or to spend the resource on such cases while others need our attention much more.  So, we do have systems through which we try to sift out all of that and to focus on the high-risk cases.

 

Your first question asked about the breakdown.  We do conduct a breakdown of the difference between customer fraud and customer error and official error.  Customer fraud makes up 0·5%, customer error makes up 0·2%, and staff error makes up 0·4%.  We break them down into those three categories and make the distinctions.

 

Mr Copeland:

Lastly, can you confirm that someone who finds themselves accused of benefit fraud enjoys the presumption of innocence until proven guilty?  How is that evidenced in the way in which those cases are approached?

 

Mr McConnell:

Every individual has the right to say that we are wrong in what we allege.  The Public Prosecution Service (PPS) is the body that decides whether, on the strength of the evidence, a case should go for prosecution, and it is then up to the courts to decide guilt or non-guilt on the basis of all of that.  We refer, as the police or others would do, a suspicion based on evidence.

 

Mr Copeland:

I thank the Chair for his indulgence.  Do you operate the same principle as the police in preparing a file by applying three tests:  the cost of a prosecution, the likely success of a prosecution and, possibly, the public interest?  Do you then send the file in all cases with a recommendation to the PPS?

 

Mr McConnell:

We do not have the power to take someone to court.

 

Mr Copeland:

I am not asking you that.  In many cases, the police can send a file prepared for criminal law with a recommendation for no prosecution, a prosecution or a change of charges.  What is the nature of the communication that takes place between you and the PPS?  Is it just a statement of the evidence with a recommendation that a fraud has been committed?

 

Mr McConnell:

Yes, that is exactly it.  We provide our evidence, which we will have gathered through the information-gathering process, to the Public Prosecution Service, and we say that, on the basis of that evidence, we consider that a fraud has been committed.  It is up to others —

 

Mr Copeland:

I understand that.  If you had investigated a file, would you send it to the PPS and say that you have investigated it and concluded that there are no grounds for a prosecution?

 

Mr McConnell:

I would need to speak to the operatives — I do not want to say the wrong thing.  As far as I know, we send the cases where we believe that the test of the standard of evidence has been met.  We offer individuals the opportunity for administrative penalties or formal cautions where we believe that the case merits that different approach.  It is for us, as an agency, to make that decision at that stage.

 

Mr Copeland:

How is that communicated back to the person?  It can happen that someone is the victim of a malicious claim.  If you investigated it and the person went through the whole process and had the stigma of investigation but did not get a letter from the PPS saying that there will be no prosecution because there is no evidence, would there be any communication from you to say sorry?

 

Mr McConnell:

Again, I would need to check, because I am not involved in the actual operation of it, but, as far as I know, we would write to the person at the end and say that it is finished.  I will certainly check that for you.

 

Mr Copeland:

I would appreciate that.  Thank you.

 

Mr Campbell:

A couple of questions arise out of the provisions in the Welfare Reform Bill.  It contains information on penalties and debt, and you mentioned that in your presentation.  Why does the Bill contain a provision to remove formal cautions?

 

Mr McConnell:

Currently, if an overpayment, for example, falls below a certain level, we say that that is not a case that is suitable for prosecution through the courts and that we will deal with it either through an administrative penalty or a formal caution.  We apply a two-tier system whereby, in some cases, we apply an internal fine — we call it an administrative penalty — and, in others, we offer the person a caution as an alternative to an administrative penalty and as an alternative to going to court.  At the GB level, the proposal in the strategy is to remove cautions as a means of dealing with suspected fraud and, rather than a three-tier system, to simply have a two-tier system of an internal fine / administrative penalty or court.  The caution would be removed as a way of dealing with cases.

 

Mr Campbell:

Does that mean, in layperson’s language, that that will eliminate any grey area about a level that might be perceived as a slight oversight of some kind but not really sufficient to proceed with?  What is the rationale for removing the formal caution?

 

Mr McConnell:

Anyone who goes forward for an administrative penalty or a caution or who goes to court does so for benefit fraud, not customer error.  I make that point very strongly.  The difference between the caution, the penalty and the court is about the means by which you deal with the fraud, but it has to be fraud.  It is more around the level or extent of the fraud involved.  For fraud that is of a lower level or smaller scale in comparison to others, as we have seen in the sentences table, we may say to ourselves that the case is more suitable for a caution.  It is still fraud — it is still intentional behaviour — but it might not be on such a scale that you want to send it to the courts in the way that you would with other cases.

 

Mr Campbell:

Right.  I think that Mickey Brady made the point that the amount of money attributable to error is close to or not dissimilar from the amount of money attributable to fraud.  Did I hear you right: did you say that both of those and their combined total are coming down?

 

Mr McConnell:

Yes.  The overall level has come down.  The overall level of all loss added together — 1·1% — has come down.  It was 1·7% when we were before the PAC in 2008.  It had been higher previously, when, I think, it was over 2%.  Yes, all have come down: customer error is down, staff error is down, and fraud is down.

 

Mr Campbell:

That is what I thought you said.  Why, then, is the rate of recovery going up?

 

Mr McConnell:

That is a separate issue.  That is about the agency recovering the overall debt stock.  Every time that we recover overpayments, it adds to the overall level of debt that is out there that needs to be paid back to the agency.

 

Mr Campbell:

I am missing something there.

 

Mr Brady:

So am I.

 

Mr Campbell:

The global amount that is due to what may be described as intra-departmental error, or problems within the system, has been declining.

 

Mr McConnell:

Yes.

 

Mr Campbell:

There is also fraudulent activity on the part of claimants, which has also been declining.

 

Mr McConnell:

Yes.

 

Mr Campbell:

The proposal is that the rate of recovery, presumably in both categories — whether people have received money fraudulently or in error — will go up.

 

Mr McConnell:

No.  This is about increasing the rate at which we recover the money itself.

 

Mr Campbell:

Yes.  So, you will recover it more quickly.

 

Mr McConnell:

Yes.

 

Mr Campbell:

Why is there the need to recover it more quickly if the overall amount is coming down?

 

Mr McConnell:

It is because the overall debt stock is still significant.  I do not have the figure, but the debt stock runs to millions of pounds that is due back —

 

Mr Campbell:

Yes, but it is less than it was three or four years ago.

 

Mr McConnell:

Yes, absolutely, because the overall level of fraud and error is down.  Nonetheless, the intention is to reduce the level of the debt stock slightly faster than that is being done at the minute.

 

Mr Campbell:

That looks to me to be a fairly significant increase for somebody who is in debt and finding things tight.  It is £3·30.  Is that right?

 

Mr McConnell:

Yes, it is £3·30 a week.

 

Mr Campbell:

That is about 25% of an increase in the recovery amount, which is a pretty significant increase for somebody who is on benefit entitlement, particularly against the backdrop of what you said about the problem reducing.  If I were a benefit claimant who was the recipient of an error on the Department’s part — not fraudulent activity on my part — and I read that the number of those errors was reducing each year, and then I was told that the amount that I was expected to repay was to increase by 25%, I might be a bit peeved.

 

Mr McConnell:

I accept your point about the tightness of money involved.  That is the maximum amount that we can recover.  We may want a slightly faster, higher level of recovery in cases where we deem it to be appropriate to do such a thing.  That is not to say that it will happen in all cases.  Obviously, cases would have to be looked at individually.  Again, the regulations will get into the detail of the added protections around the ability to repay and so on.  We are simply saying, at a very high level, that the maximum in whatever number of cases may increase but not in all cases.

 

Mr Campbell:

Right.  A final question on that:  do I take it from your comment just now that the maximum increase of a significant amount would be levied on the more extreme offender to try to ensure that the debt is reduced more rapidly?

 

Mr McConnell:

We have not got to that stage of development.  That will come in the Bill and the regulations.  This is about saying that we want to widen the power to increase the maximum amount, where applicable.  We will have to get into all that detail as we go through the process.

 

The Chairperson:

Is it fair to say that, while the levels of fraud and customer and departmental error are going down, the thrust of this, as you outlined in your opening remarks, is about introducing a new and much more rigorous approach to all that?  As Gregory put it in his question, the figures are going down but you want to increase the recovery levels.  You also mentioned formal cautions.  A formal caution is actually a sanction.

 

Mr McConnell:

Yes.

 

The Chairperson:

But you want to remove that.

 

Mr McConnell:

Yes.

 

The Chairperson:

I use that example to illustrate that you are trying to introduce a much more rigorous and, I would say, probably more punitive environment.  A formal caution represents a sentence.  You can go through the courts and get a discharge or a warning; it is an actual sanction.  People may argue that it is either stronger or lighter, but I would have thought that it would still be a useful device, because there are grey areas where a formal caution may be appropriate.  At least you are putting a person on notice.  That happens in magistrates’ courts every day of the week.  I am just making that point.  I am trying to rationalise why, on one hand, figures are going down and problems are seen to be decreasing but, by the same token, the rate of recovery is going up.

 

Mr Douglas:

Thank you for your presentation.  I want to go back to Gregory’s point.  I think that we need to revisit the recovery aspect.  In the case of an overpayment, someone is either innocent or they are not.  Conrad, you said that you would look at the appropriateness of the whole thing.  I cannot see how, if someone is innocent, they get 25% taken off their benefit, and someone else who is innocent has it reduced by 15%.  That is something that we need to look at.  You are either innocent or you are not.  There are no continuums or levels.

 

Secondly, I found it difficult to read some of this stuff.  Most of it is OK, but I need to put my glasses on to read a couple of the slides.

 

The Chairperson:

Sammy, it is not the Department’s fault.  We have been trying to dispense with PowerPoint presentations.  It is death by PowerPoint presentations, sometimes.  It is not the fault of the officials.

 

Mr Campbell:

Sammy wanted glasses, but I think that he needs binoculars.

 

Mr Douglas:

Conrad, you talked about the fraud hotline.  Can you explain that a wee bit more?  I assume that it is different from the PSNI hotline.  If someone reports an incident, the PSNI has a duty to investigate it.  You say that you do not do that; you sift through it.  Will that still be the case?

 

Mr McConnell:

Yes.  We will always have referrals.  We will still have a hotline.  At present, the problem is that there are a number of phone numbers for hotlines run by different organisations.  The idea is to make things simpler for people by having one number to ring for any type of benefit fraud, whether it involves tax credits, housing benefit or SSA benefits.

 

Mr Douglas:

Do you intend to continue to sift out, though?

 

Mr McConnell:

Yes.  We will always want and need to sift out referrals, because we simply would not have the resources, nor would it be value for money, to investigate thoroughly every case.  Fraud investigations, for benefit fraud in particular, are very complex and involve all sorts of history and statements of income and so forth.  They are quite deep.  Therefore, we need to make sure that, where we invest money and time in a fraud investigation, we are doing so on high-risk cases.  We want to move referrals that may be scurrilous or weak or do not seem to point to very much to one side as quickly as we can.

 

Mr Douglas:

In all our constituencies, we have neighbourhood disputes in which people may want to make malicious allegations.  Do you have any idea of what percentage of allegations are sifted out?

 

Mr McConnell:

I am not sure.  I would need to check.  Ultimately, we will always have to sift on the basis of our resource.  We have x number of fraud investigators, and each case takes, on average, a certain length of time to investigate.  Irrespective of the level of referrals, if our resource is fixed and that is what the budget allows, we will always have to draw a line that gives us the capability of dealing with x number of cases based on x number of investigators.

 

Mr Douglas:

I understand.  Thank you.

 

Mrs Cochrane:

Thank you for your presentation.  I do not think that any members have commented that it is good to see that the figure has come down from 1·7% to 1·1%.  I want to start by saying that. 

 

Tommy O'Reilly’s briefing paper refers to there being similar fraud and error provisions.  Are the penalties and sanctions referred to the same as those in GB, and are we bound by parity to implement them?  Is there any breathing space?

 

Mr McConnell:

We have legislative colleagues, Anne McCleary and others, who are much more in tune with those issues.  I know that parity is an issue, and there are questions over whether we must act in certain ways.  A general point is that, if we do not take every opportunity to reduce loss within the benefits system, it is an issue.  I will refer the issue of pure parity to colleagues who know about the legislation.  We can write to you about all that, if that is acceptable. 

 

Mr Brady:

I have a comment that follows on from what Gregory said about one side going down and the other going up.  The bottom line is that, if someone is on, for example, income-based jobseeker’s allowance at the moment — presumably, they will go on to universal credit eventually — by the Government’s own admission, that is subsistence level.  That has always been said, from the days of supplementary benefit right through.  People’s income will then go an extra £3·30 below subsistence level, because that money will be taken out of their benefit.  The point that Sammy made is also something that needs to be looked at.  Benefit levels are not enough to warrant that kind of amount being taken.  From my experience of many cases over the years, you had to contact the Department and say that it was causing severe financial hardship.  They would then look at each case individually, but the amount was not always reduced.  I think that this will cause more problems than it is intended to solve.

 

The Chairperson:

It is my intuition that, if you increase the scope for reclaiming money from a person who innocently received an overpayment, that takes away an incentive for the Department to continue its work to drive out its errors.  Has any consideration been given to that?  Does deciding to increase the recovery rate through stronger sanctions not disincentivise the Department? 

 

Mr McConnell:

I suggest that it does not.  We have a system that measures the levels of fraud and error in the benefits system.  It is independent of anything else in the agency, and it reports its figures to the Audit Office and everybody else.  How much of that loss we are recovering is a separate figure and a separate issue.  It is discussed separately in the annual report and the accounts and in the Audit Office’s opinions on all of that.  It does not impact on the overall level of loss — 1·1% or 1·3% or whatever it is.  That loss will always be that loss, irrespective of how quickly we recover the overpayments from it.

 

The Chairperson:

OK.  Thank you, Conrad.

 

Mr Durkan:

Thank you for the presentation.  Is the loss that you have been talking about net of or measured against unclaimed benefits?  You gave the figure of £48 million — £67 million including housing benefit — as the loss in 2010.  What was the sum of unclaimed benefits for the same time?  Is there a strategy to tackle that?  Is it being approached with the same zeal as this?

 

Mr McConnell:

There are two issues.  Loss is the amount of incorrect benefit that we estimate has gone out and has been paid.  We have underpayments as well, which is the benefit that we think is incorrect but has not been paid.  Individually, the figures are 1·1% loss for the agency and 0·5% underpayment.  That means that the overall incorrectness, if you add everything together, is 1·6%. 

 

Underpayment relates to the issue of what we are doing to make sure that people get the right amount of money.  We would always go back to the core point that we are really about trying to get the money right, irrespective of whether it is over or under.  For example, the people who we fund to tackle error, in particular, find that about half of error is underpayments.  It is not all about loss and reducing the overall level of money going into society in error; it is about getting the money right.  Underpayment is as key for us as anything else.  It is a major outcome of our error activity, and it also ties into the agency’s wider benefit uptake strategy.  For example, we are doing much more this year around trying to encourage benefit uptake, with television campaigns, the outreach programme and targeted exercises aimed at individual groups of people.  We take the view that we do a lot on fraud activity, but we also do a lot to tackle error, and much of that is about underpayments as well.

 

Mr F McCann:

To pick up on that point:  it is my understanding that, if you are underpaid and try to claim the money back, you will go back only three months.  However, if you are overpaid, there is no time limit.

 

Mr McConnell:

I would need to check that.  I am not sure.

 

Mr F McCann:

I am just raising the issue of the level of unfairness that there is about that.

 

Mr McConnell:

I am not sure.  I will check whether that is right.

 

Mr F McCann:

Thank you.

 

Mr Brady:

To make a final point:  Mark talked about people not claiming.  When you talk about underpayment, are you talking about people who are claiming benefit but are not paid the proper amount?

 

Mr McConnell:

Yes.

 

Mr Brady:

I think that Mark was referring to the fact that approximately £1·9 million a week in pension credit is going unclaimed.  That does not go back into a pot that is used for other things in the North; it goes back to the Treasury.  So, we are talking about two different things.  Underpayment relates to people who have claimed but are not being paid the proper amount.  The campaign that you talked about is to encourage people to claim.  Approximately £1·9 million a week in pension credit is unclaimed.  You are talking about almost £100 million a year.  Set that against the amount involved in alleged fraud and error, and it is a lot more.

 

Mr McConnell:

The campaign that I referred to was exactly about that problem.  The television campaign and the outreach programmes are not about underpayments.  They are about people not claiming when they should be.

 

Mr Brady:

I just want to make sure that people know that there is that distinction, because there is a distinction there.

 

The Chairperson:

OK.  We have had quite an exchange.  I thank John and Conrad for their presentation and for dealing with the questions that were put.  I have no doubt that we will return to this issue.

Find MLAs

Find your MLAs

Locate MLAs

Search

News and Media Centre

Visit the News and Media Centre

Read press releases, watch live and archived video

Find out more

Follow the Assembly

Follow the Assembly on our social media channels

Keep up-to-date with the Assembly

Find out more

Useful Contacts

Contact us

Contacts for different parts of the Assembly

Contact Us