I beg to move amendment No. 51, in page 11, line 35, after 'proceedings', insert

`where benefit is in payment or, if benefit is not in payment, when the entitlement first arises,'.

With this it will be convenient to take amendment No. 52, in page 11, line 37, at end insert—

`(6A) The Secretary of State shall provide for a central record to be maintained of

(a) all convictions for benefit offences

(b) all subsequent periods of disqualification of sanctionable benefits

(6B) Such records shall also include reference to the National Insurance Number held by the individual concerned.'.

I thought that considering the amendments together would be easier for the Committee, because amendment No. 52 is consequential on amendment No. 51.

The purpose of amendment No. 51 is to include the sanction implied in page 29, paragraph 106 of the explanatory notes to the Bill, which state:

``The offence for which the second set of proceedings has been brought must occur within a period of three years of the date on which the offender was convicted for the first offence.''

If such a hazard exists, it should be dealt with by the Bill. If the two-strikes provision is to be effective as a deterrent, it must be clearly enforced.

Amendment No. 52 sets out the measures that should be consequent on amendment No. 51. The most important factor is the use of fraudulent national insurance numbers, which are known to most of us as NINOs. That matter is dealt with in the second part of amendment No. 52, which would also provide for the keeping of a clear and precise record of offences that would enable an enforcing officer at the Benefits Agency or the Department to ensure that the claimant served any period of disqualification. The loss of benefit provision must be properly enforced to provide the deterrent that the Government seek.

Amendment No. 52 reflects the concern that we all have about the large number of national insurance numbers in operation. There are more than 81 million such numbers, and the Minister of State has helped us by breaking down where they come from. We understand that every person in the country, whether they work or not, has a national insurance number. Therefore, that accounts for about 60 million. Roughly 12 million NINOs remain in existence that belong to people who have died whose estates may affect other people's entitlement to benefit. That makes roughly 72 million, so about 9 million are going begging.

In a reply to a question about NINOs asked by the right hon. Member for Birkenhead (Mr. Field), the Minister of State said that

``only in 11 cases has it been established that the account was deliberately created in order to establish a false identity and fraudulently gain access to benefits.''—[Official Report, 13 March 2000; Vol. 346, c. 53W.]

The Balham experiment, which showed that there were about 220,000 mismatched numbers, resulted in about 166 arrests and six deportations for identity fraud. In a debate on the Bill in the other place, Baroness Hollis said that extending the experiment to neighbouring Benefits Agency offices in 1999 resulted in a further 46 arrests. The Balham experiment suggests that the problem with NINOs is much larger than the Department found when it made a computer check. I understand that the computer check is being run again. Is the Minister of State planning to extrapolate the Balham experiment to other benefit offices so that we can get a feel of whether the 9 million missing NINOs is made up of the 220,000 mismatched numbers multiplied by the number of benefit agency checks required? That may give us some idea of what percentage of those 9 million NINOs are fraudulent and, therefore, how tough the Department will need to be to sort out the national insurance numbers system.

Taking the two amendments together has made it easier for the Committee to debate the problem of benefit fraud. I look forward to the Minister of State's response.

I think that I can satisfy the hon. Lady--I cannot accept the amendment, but I can do the next best thing. I shall not read my speaking notes, but shall use the pre-note. I am told that amendment No. 52 would have no effect; what it states is precisely what the Department intends to do. Regulations will be made once the Bill is enacted. I shall not speak about the technicalities of amendment No. 51. The argument is flawed, but that is inevitable because of the way in which such measures are drafted. However, such a measure is not needed.

We intend to prescribe in regulations made under the powers in clause 7(6) that the disqualification period will start either within 28 days of the second conviction if, at that point, the person is entitled to a sanctionable benefit or, if the offender is not entitled to a sanctionable benefit at the time of the second conviction, when he first becomes entitled to a sanctionable benefit within three years.

I shall try to be as brief as I can—I could be a lot more wordy if I used my speaking notes. We can deal with the points raised by the hon. Member for Beckenham. First, I have outlined how we intend to operate. Secondly, that will be made clear in regulations. There will be no doubt about when the ``two strikes and you're out'' provision will be used, and I repeat that it cannot be used unless there has been a second conviction. The Department will not operate on the basis of punches; it will require convictions in a court of law before the ``two strikes and you're out'' sanctions can be applied.

There are not 9 million missing NINOs or missing people. Without the benefit of my notes, I can tell the hon. Lady that the figure of 60 million, to which she referred, plus the 12.5 million numbers relating to deceased people, which are kept on the system, make up the 72.5 million. That leaves another 9 million to make up the 81 million. Since the national insurance system started, anyone who has worked here or who was born here and left the country permanently has had a NINO. Those numbers still exist.

Anyone who came to the country for a couple of weeks' fruit picking in East Anglia or in Scotland was allocated a NINO. They might never come back to the country, but they have been given a number. Millions of people have left the country since 1948 and, over those 50 years, others came to work here temporarily. They were given a NINO, but they may have left and might not come back. They amount to a substantial number of people. I am not saying that that represents the grand total, but it is substantial. There are not 9 million missing people or missing NINOs.

As to data cleansing, on Second Reading I said that in the last year for which we have figures, we cleansed and removed more NINOs than we did in the previous few years combined. However, I do not have the exact number. During the cleansing operations, we have found instances of either one person with two NINOs or two people sharing the same one. Such things happen, perhaps because an employer supplies the wrong number or because of human error during data inputting, but discovering such discrepancies is part of the cleansing operation; then we can upgrade the system.

The NINO system is an actively managed operation. It is not a static list of 80-odd million numbers, locked away in the Department's computers and left there. Where there are doubts or suspicions, flags are placed on numbers so that if they are ever activated again, the system will show that that has happened and that we need to investigate why and by whom the number is being used.

The hon. Member for Beckenham referred to the Balham project, which was a pilot. We have put it on record that the Balham techniques of in-depth analysis and more systematic questioning about NINOs are being rolled out across the country, virtually as we speak. They will be fully rolled out by October, so there will be designated people across the country trained in those techniques. I have sat behind an interviewer at Balham who was going through the process and I know that we shall apply what we learned at Balham across the country. That is not to say that we can in any way extrapolate the Balham arrests and deportations. However, we can say that those who were interviewed in Balham about national insurance numbers were spoken to on the basis that they had a job offer, but did not have a NINO.

That situation can apply only under a peculiar set of circumstances. It does not occur in the case of 16-year-olds born in this country, because their NINOs are activated automatically. In the light of the Balham project, the issue of national insurance numbers will be a much more tightly controlled and systematic operation. The fruits of that will become clear as the months pass.

I have been as brief as possible, and I hope that I have answered the hon. Lady's points about the amendments and about the Balham project. However, I am happy to answer any other question that she has.

I am grateful to the right hon. Gentleman for saying that our amendment was correct. It seems unfair that it should not be included in the Bill, but no doubt a version of it will be included in the regulations. Will those regulations be subject to the affirmative resolution procedure? If so, I am sure that we will meet in this Committee Room or another to thrash out the question in due course.

It is nice to know that we are on the right track. I am sure that the right hon. Gentleman or his successor will keep us and the country abreast of progress on the cleansing of the NINO records. That is crucial—I have constituency cases that are affected by the problem.

I can confirm that the regulations will be subject to the affirmative resolution procedure.

That is the best news that we have heard today. On that basis, I beg to ask to leave to withdraw the amendment.

Amendment, by leave, withdrawn.

I beg to move amendment No. 53, in page 11, line 44, at end insert—

`; likewise any third party either informed of the restriction or contacted as part of the investigatory process shall be notified of the quashing of the conviction.'.

With this it will be convenient to take amendment No. 55, in clause 9, page 15, line 17, at end insert—

`; likewise, any third party either informed of the restriction or contacted as part of the investigatory process shall be notified of the quashing of the conviction.'.

The amendment is designed to provide a system of redress for a claimant whose benefit is withdrawn. We acknowledge that such a system already exists, under which, when a conviction that has resulted in the restriction of benefit is quashed, all payments and adjustments are made to the claimant as though the conviction had not occurred. However, we are concerned about those who are subject to an investigation that is found to have been ill advised and unnecessary. Such abortive investigations have an effect. Most of us will know of people who have been turned down for credit or hire purchase because the credit reference agency did not have sufficient information. That can happen because the person is not on the electoral register. It happened to me when I lived in the city of London; the city of London electoral register was not on the credit reference agency's computer and I was turned down for credit. Luckily, I could use a cheque, but not everybody can do that.

If an investigation proves to be unfounded and is abandoned, there may still be traces—footprints—on a person's record showing that such an investigation has taken place. We want to ensure that there are safeguards for people who find themselves in that position. I should be grateful if the Under-Secretary would reassure the Committee on that.

I hope that I can reassure the hon. Lady. The clause is designed to ensure that any money resulting from withdrawal or reduction of a claimant's benefit as a result of a two-strikes sanction is repaid if one of the convictions is quashed, which is right and fair. The same principle is applied to other benefit sanctions.

Accepting the amendments would mean that the Department would have to inform any third party contacted over the course of a fraud investigation, who may have been notified of the imposition of a sanction, that the conviction had been subsequently quashed on appeal. That would mean that any individual, agency, employer or Government organisation—irrespective of the extent of their involvement in the investigation—would have to be notified of the quashing of the conviction whether they knew the outcome of the original investigation or not.

It has never been normal practice to go back through an investigation file and inform third parties either of the result of an investigation in which they may have been involved, or whether a sanction has been imposed. The process that the hon. Member for Beckenham suggests should be adopted would be cumbersome and expensive. I hope that the key thing she is getting at with the amendment is that any quashed conviction ought to result in the repayment of benefit rather than a mass declaration of innocence to all the people or agencies that may have been involved in the initial investigation.

Mrs. Lait rose—

The hon. Lady is champing at the bit, so I shall give way.

I am talking not about people who have been convicted, but people who have been investigated and found to have no case to answer.

I thank the hon. Lady for clarifying that. I do not know whether it will reassure her—I hope that it will—to know that we are making arrangements with credit reference agencies so that footprints will not be left in any records and credit reference agencies will not record criminal convictions. No footprints should be left by any of our investigations. I hope that that will persuade the hon. Lady that her worries on this point are groundless, and that she will consider withdrawing the amendment.

I shall be happy to withdraw if the Under-Secretary assures me that if evidence emerges of any footprints remaining—not just with credit reference agencies but with utilities, banks, insurance companies or any of the other organisations that the Department will be able to contact to check people's bona fides—the Government will take immediate steps to ensure that they are eliminated and that the necessary qualifications are placed on people's records.

It is not our aim to leave footprints and place innocent people in difficulties with their credit reference agencies or in other business transactions. That understanding will be an important part of the agreement when we share data. We do not intend the Bill to result in innocent people being put on a blacklist because of information sharing, whether that involves utility files—which the hon. Lady should remember contain addresses, not names—or any others. If anything were to go wrong with our proposal, it would be incumbent on those involved in the information sharing to rectify matters as quickly as possible, but we hope that such errors will be preventable.

I am grateful for the Under-Secretary's reassurances, and I accept them. I am glad to have it on the record that the Government will take instant action if anything is proved to have gone wrong, and I hope that in such an event the Under-Secretary would share the information with the House. I beg to ask leave to withdraw the amendment.

Amendment, by leave, withdrawn.

I beg to move amendment No. 54, in page 12, line 17, leave out `(ii) working families' tax credit'.

With this it will be convenient to take amendment No. 44, in page 12, line 19, leave out from `pay;' to end of line 20.

The amendment is an attempt to include working families tax credit as a payment which, if the subject of an offence, could be withdrawn or lost. We have already discussed working families tax credit as a Treasury issue, and the fact that the authorised officer is at a higher level, under the Taxes Management Act 1970, than the relevant officer in the Department of Social Security. However, all the international evidence is that tax credit fraud is quite prevalent and that there are potential problems.

Amendment No. 44 is intended to tease out why the Department has chosen to exclude retirement pension from the list of disqualifying benefits but to include war pension. When the Bill was debated in another place, three Lords expressed concern, on behalf of war pensioners and their organisations, about the inclusion of war pension on the list. When Baroness Hollis responded to the debate, she said:

``So far as I am aware, we have detected only six frauds during the past six months involving war pensions.''—[Official Report, House of Lords, 16 January 2001; Vol. 620, c. 1091.]

Given the scale of that fraud, one assumes that there were no repeat occurrences.

This part of the Bill will probably have no effect, so why should it not be taken out? It should be removed, because to leave it in would offend the war pensioner community. I cannot find the exact reference, but I am sure that, when challenged as to why retirement pension was not included on the list, Baroness Hollis said, ``Because we do not think there is a problem of fraud with retirement pensions.'' Why exclude retirement pension but not war pension?

The Minister of State, in responding to the Second Reading debate, seemed to imply that the logic of my argument required that retirement pension should be included. As he well knew, that was not the point that I was making. I was suggesting that if there is a good case for excluding retirement pension, because there is practically no evidence of fraud, there is a good case for excluding war pension, too. It is on the record that the number of cases of war pension fraud is negligible, so the provision is unlikely to have any practical effect, except that of giving offence to Britain's war pensioners. On that basis, I hope that the Government will be sympathetic to the amendment.

I am sorry if I misunderstood the hon. Gentleman on Second Reading. I do not have a window into people's minds, and I naturally assumed that he was arguing the other way around.

It may be repetitive, but it is important to reiterate what was said in the House of Lords. There is a distinction between retirement pension and war pension, and between retirement pension and many other benefits. It is almost impossible for someone to commit fraud with with regard to retirement pension. As long as people are the right age, are who they say they are, and have paid their national insurance contributions, retirement pension is theirs by right. There are no qualifying conditions. It is simple—date of birth and identity can be checked easily.

Fraud can occur when the person entitled to the retirement pension no longer requires it, but someone else continues to collect it. Sometimes the centenary lady at the DSS spots that; sometimes it goes on for a long time. In such cases, the fraud is committed by someone entirely different, not by the benefit claimant. Retirement pension fraud committed by benefit claimants themselves is almost non-existent. That is one reason for not including retirement pension in the Bill.

As the hon. Gentleman told us, Baroness Hollis said that in the past six months only six frauds involving war pensions had been detected. The level of war pension fraud has never been measured. Information is available only on the frauds detected. No benefit review, like those of jobseeker's allowance and other benefits, has been conducted.

The hon. Gentleman has asked me to distinguish between war pension and retirement pension. Unlike retirement pension, war pension has areas in which fraud can occur. This is most likely in five areas. With treatment allowance, false claims about attendance for treatment can be made. Fraud in respect of the allowance for the lowered standard of occupation takes the form of a failure to declare an increase in earnings. There could also be a failure to declare employment when an unemployability supplement was being paid. False statements could be made about walking ability when war pensioner's mobility supplement was being paid. There could be a failure to declare co-habitation when war widow's pension was being paid.

Rigid conditions covering age, identity and national insurance contributions apply to retirement pension, but other factors apply to war pension. I am not casting a slur on war pensioners, but fraud has occurred, and I am not aware of any fraud concerning retirement pension, except when it is committed not by the pensioner, but by someone else who has collected the pension after the pensioner's death.

There is a difference between the two pensions. Conditions covering war pension and some of its supplements do not cover retirement pension. I am referring to the basic retirement pension, not the add-ons, the minimum income guarantee and so on, so the scope for fraud almost non-existent--I am not daft enough to say that it is totally non-existent. On the other hand, there is scope for fraud in war pension.

I accept the Minister of State's point that war pension is more complicated and that there is more scope for abuse, but the provision would apply only when fraud had been confirmed—and there have been only six cases in six months. The likelihood of a conviction happening twice is probably nil. The potential for fraud does not matter, because the provision would be activated only if fraud was detected. The Minister's figures show that it was detected only half a dozen times, and the chance of it happening twice must be next to nil, so why not remove the provision from the Bill?

It is even worse than that. I may be undermining my case, but I want to be as open and frank as possible: six cases of fraud were detected, but there were no prosecutions. Judgments were made on the circumstances of each case, and no one was prosecuted. It can be argued that the provision should be removed, but I have listed five factors with potential for fraud, which show that there is a great difference between war pension and retirement pension, so it is sensible to leave war pension on the list, but not to include retirement pension.

This is not a question of promising or planning to change policy on prosecutions. There is no policy not to prosecute; there happened to be no prosecutions, but that was not a decision for Ministers. War pension includes factors--mobility, earnings, and cohabitation, for example--in connection with which fraud could occur, but those factors do not apply to retirement pension, and we do not want to wipe out the sanction, so it is sensible to leave war pension on the list. We constantly review legislation, and the House will also examine what happens in the fullness of time, but that is for the future, not now. I have explained the difference between retirement pension and war pension, and there are good grounds for leaving war pension on the list.

The hon. Member for Beckenham did not press the point about the working families tax credit, which we debated at an earlier sitting. It is available only to people who are working at least 16 hours a week, and replaces family credit. If a sanction were applied--I am not saying that we have considered that, and if we did, it would be a matter for the Inland Revenue, not the Department of Social Security--it might hit the children and other members of the claimant's family, and we have sought not to do that in respect of other sanctions. It might put a person's employment at risk, which is not what the working families tax credit is intended to do. The idea is to make work pay, and to get people off benefits and into work. The working families tax credit was designed to ensure that benefits paid in respect of a child or a partner are not sanctioned.

As I have said, tax credit fraud is a matter for the Government as a whole, and for the Inland Revenue and the Treasury in particular. In saying that, I am not opting out; the Government are actively interested in that issue. The credit—I almost said the benefit, but I stopped myself—is new, and time will tell how we will ensure that payment is made to the correct people at the appropriate rate, and prevent fraud in the system.

Perhaps I will never be able to convince hon. Members about war pensions—an issue with an emotive aspect that I understand and respect. If I were sitting where the hon. Member for Northavon is sitting, I would be making the same case that he is making. However, having examined the matter I believe that there is a genuine case for including war pensions in the list, and I am quite comfortable with its inclusion. In terms of the rules, there is a distinction to be drawn between a war pension and a retirement pension. I am not simply spouting that line because I am taking the Queen's shilling, as it were. The distinction between a war pension and a retirement pension justifies including the former in the list, and it certainly justifies excluding the latter.

I am grateful to the Minister for his open response, which, as he hinted, has in fact strengthened my argument. He is arguing that a retirement pension and a war pension are different forms of benefit, and I fully accept that. However, the practical question is whether they are different in terms of fraud convictions, because it is only such convictions that will trigger the provisions in the Bill. He said that there had been no war pension fraud convictions in the past six months, but seemed unsure whether there had been any retirement pension fraud convictions involving the claimant.

There are none that we are aware of.

If that is so, the two benefits have that in common. Given that the war pension is more complicated, it could in theory give rise to prosecutions, but the reality is that in the past six months it has not.

May I make one further point? We are not making a big issue of the ``two strikes and you're out'' principle. We have made it clear that it is a deterrent, and that we do not expect many cases to arise. Off the top of my head, I think that of the millions of claims made, we anticipate only 500 cases a year. Rather than dwelling on the huge problems that exist, we are trying to send a signal that will persuade people to change their behaviour. We accepted the recommendation, laid out in the Grabiner report, that such a signal would be useful and would act as a deterrent. As I said, we do not expect thousands of cases, but it is important to make clear the existence of a deterrent. If no prosecutions arise, that will be the big success. We want not to sanction people but to prevent them from committing benefit fraud.

The Minister says that the point of the provision is to send a signal, but to whom will it be sent? We are discussing war pensioners, and it is the suggestion that serial fraud among war pensioners is a real issue that has given offence. The provision will send the signal to war pensioners and their organisations that, in the opinion of the DSS, serial war pension fraud is a real, or potential, issue—but the evidence does not bear that out. There has been not a single conviction in the past six months, let alone a repeat one.

The Minister will be unsurprised to hear that I do not find his response convincing—and perhaps he himself does not. I am not sure whether I am at liberty to divide the Committee on this issue, but I hope that the House will be able to return to it in future.

We have debated the working families tax credit before, so I shall be happy to withdraw the amendment.

The hon. Gentleman could press amendment No. 44 if he wished to, but I take it that he does not.

Mr. Webb indicated assent.

I beg to ask leave to withdraw the amendment.

Amendment, by leave, withdrawn.

Question proposed, That the clause stand part of the Bill.

I want to use the discussion of clause 7 as an excuse to ask the Minister a question, which may also apply to clause 8. In a decision to reduce benefits, what discretion is available to Benefits Agency staff? In particular, jobseeker's allowance can be reduced rather than removed entirely, which, as mentioned in the explanatory notes, passports the claimant back on to housing benefit and council tax rebate. I want to discover whether discretion is involved in that process. Does it involve withdrawing one of the disqualifying benefits? I understand that maternity allowance cannot be withdrawn from a claimant. Are there circumstances in which certain allowances can be withdrawn? I ask that because there are many serial offenders—some in well organised criminal gangs—who commit fraud. It would be nonsense for such people, who have prosperous lifestyles and are claiming everything that they can get, including housing benefit and maternity allowance, to be allowed to keep those allowances when they had been convicted of a serious fraud offence. Will the hands of Benefits Agency staff be tied by the regulations?

If there is one clause that we would remove from the Bill, it is this one. It allows the Department to sanction someone who twice has been convicted of benefit fraud, and people can be sanctioned on benefits such as the means-tested safety net guaranteeing the basic minimum income. We are not arguing about whether someone who has committed fraud twice is in the wrong, but about whether the punishment under clause 7 is appropriate. In all cases of punishment, we must consider the concept of appropriateness and whether we have the right level of punishment for the offence. If society deems that someone has committed a serious offence, they are deprived of their liberty; we do not deprive them of food, clothing and shelter, because we do not believe that is an appropriate punishment even for the most serious crimes. Even murderers are fed, housed and clothed. Our contention is that the consequences of clause 7 would leave some of those thereby sanctioned below the poverty line and potentially unable to feed, clothe or shelter themselves. That is why we object to the clause.

I recently raised that issue with the Department during a Committee on a statutory instrument. The Under-Secretary responded by saying that the breach of community service orders was another action for which the Government intend to introduce benefit sanctions. My noble Friends in the other place, talking about the Government's strategy of sanctions, asked about the lack of research undertaken by the Department. There has also been a failure to take account of the limited research that has been undertaken. When I challenged the Under-Secretary about that research, she prevaricated until she was informed that a research report had been published in November 1998.

I did not prevaricate; I had the research document in front of me.

I gained the impression that the hon. Lady did not have the document then, but I apologise if I was wrong.

When the Department was challenged to produce evidence of the effect of sanctions on claimants, it cited a piece of research two and a half years old. Naturally enough, I have read that report, which refers to a sample of 30 claimants. It is in the nature of qualitative research that samples are not huge, but the Department is not making much of an effort if the principal research evidence that it can adduce on the effect of sanctions on recipients is based on a sample of 30 people taken two and a half years ago.

I make that point because the report's findings do not support the Department's sanctions policy. For the record, I am referring to research report No. 86 from the DFEE and the DSS, ``Jobseeker's Allowance Evaluation: Qualitative Research on Disallowed and Sanctioned Claimants. Phase Two: After Jobseeker's Allowance''. Page 27 of that report states:

``Seven of the respondents in the first round''— that is, about a quarter—

``lived with their partner and child(ren). As might be expected, they appeared to be the hardest hit by the cuts in their benefit income.''

I am sure that the Minister of State will refer to the hardship payment system, which is supposed to mitigate the effects of sanctions—but the report examines the actual effects of actual sanctions. Families with children are the sort of people who might be entitled to hardship payments, but the research found that families with children

``appeared to be the hardest hit.''

That is not sanctioning the claimant; it is affecting the whole family, including innocent children. I should like to know how the Under-Secretary can tell us that the Department has taken account of its research in framing clause 7, when the clause goes no further than existing protections against hardship, which, according to its research, hit families with children the hardest.

A second group of respondents were those who lived alone. The report states:

``Most of the respondents who lived alone appeared not to be able to turn to their parents/family for support.''

A natural argument is, ``We sanctioned somebody, but they can go home to mum and dad, or ask their mates.'' However, the research states that that is not the case. The report continues:

``They gave the impression they were on their own and said that they could ask for only a limited amount of help from friends.''

The idea that they can get round the sanctions by asking their mates for a loan is unrealistic, and is not what the Department's research suggests.

There is other research on the effects of sanctions, which was commissioned not by the Department, but by Janet Allbeson, who has links with citizens advice bureaux. She—this quote is reproduced in the Department's research report—describes the severity of sanctions as

``causing a level of destitution out of all proportion to the `offences' involved.''

The evidence from that study reinforces the critical point that sanctions impose real hardship. [Interruption.] The hon. Member for Hexham (Mr. Atkinson) says that that is the idea of sanctions, but proportionality must be the key to punishment. We deprive people whom commit the most serious offences of their liberty, but we do not deprive them of the basic necessities of life such as food, clothing and shelter.

The study discusses someone in the community whom we would all—including the DSS—agree has committed a less serious offence, but is facing destitution

``out of all proportion to the offences involved.''

We are not arguing that there should be no punishment, but we are asking whether that is an appropriate level of punishment.

It is important that we make a stand on clause 7. It is another step on a road down which the Government have been marching, and we shall see them make more strides down that road if they win the next general election. Benefit entitlement will be seen as more conditional. Basic money for food, clothing and shelter will become conditional on jumping through a whole lot of hoops that will get tighter and tighter. The Government are clearly moving in that direction, and clause 7 takes us further down that road.

The Government may say that they do not have to worry because there are hardship provisions. However, hardship provisions still leave people below basic benefit levels, although not as far below as they might otherwise have been. Given that people on basic benefits are often in debt—to the social fund and all sorts of other lenders—and are already getting by on less than basic benefit levels, even a hardship scheme that modifies the penalty leaves them prone to destitution. Is that the punishment that the Government think appropriate? It will not make murderers destitute; it will make benefit fraudsters destitute.

If the hardship scheme is to be a defence, it must work properly, yet the Department's research report questions whether that is so. Page 28 states:

``One young man said that not only was he refused hardship payments, he was told he could not have a Social Fund loan because the sanction was intended to cause him hardship, as a punishment''.

That advice from the Benefits Agency was inaccurate. The hardship scheme leaves people below the basic benefit levels for food, clothing and shelter and, moreover, the people who run the system apparently do not know how it works.

The Government's response to such criticism from Liberal Democrats is, ``Well, people shouldn't do it, should they?'' I hear that from Ministers all the time. However, we are not questioning the fact that such acts are wrong, criminal and deserving of punishment, but whether the punishment is right. Obviously, people should not do such things. They should not murder, but they do, and we must decide on an appropriate punishment. The same is true of benefit sanctions.

We will object to the sanction on principle until two conditions are met. First, decent evidence of its effects on the people who are sanctioned must be produced to determine whether it is appropriate or proportionate. That is frequently mentioned by Earl Russell in another place, where his views are treated with respect. Secondly, when that research is commissioned, there must be evidence that the Department is taking a blind bit of notice of it. So far, it has roundly ignored such research as has been undertaken. The punishment is not appropriate to the offence, and we therefore do not accept that clause 7 should stand part of the Bill.

The hon. Gentleman is on the borderline—and has perhaps gone over it—of describing persistent benefit cheats as lovable old rogues. They are not. I have sometimes stood at the Bar of the House in the other place listening to Earl Russell and thinking, ``How out of touch can you be?'' There is nothing personal about that—I have never had a conversation with him—but he is not living in the real world.

I freely admit that the sanction is tough; it is intended to be. It is supposed to act as a deterrent to prevent people from becoming persistent benefit cheats. On any reasonable test, we have built in safeguards for some benefits. In other cases, the subject of the sanction will not be the benefit that is cheated on, but a benefit from the list of sanctionable benefits. It is a two-stage process.

The hon. Gentleman mentioned research. We assume that offenders can read. They have already cheated the system, so they must have known how to fill in a form: how to read the questions and lie in response. The first time that someone is convicted by a court of cheating the benefits system, we will put a note into their hands saying, ``If you do this again and get convicted, you're for the high jump, sonny: you'll lose some of your benefit''—assuming that it is a sanctionable benefit. There will be no misunderstanding. The person will not be able to say, ``Oh, I didn't understand; I thought that I would be able to carry on cheating the system. They never explained to me that if I did it again and was taken to court, I would lose money.'' In the past, they might have got away with it—but now, no way.

We must make it absolutely clear that there is a dividing line between us and the Lib-Dems, with their wishy-washy policies. The hon. Gentleman gave the impression that they want to defend benefit cheats. We will not allow cheats up and down the country to make the excuse that they did not understand the consequences of cheating a second time. They will not have that argument, because we will tell them the first time, ``Do it again and bang—we're going to hit you.'' No amount of past or future research can let such cheats off the hook.

As I understand the findings of the jobseeker's allowance research, a majority of respondents said that they would take care and be more wary in future, and that they had become more determined to find work. That is part of the exercise. We aim to get everybody in the country into gainful employment, if physically possible. There is no excuse not to be gainfully employed. The greatest asset that any country has is its people's capacity and willingness to work. If we fail to use any part of that asset, it is bad for us economically as well as bad for the quality of life of the people concerned. That is why we ask people what they can do, rather than telling them what they cannot do. We are trying to change the culture.

I want to spell the matter out as it is, so that there is no misunderstanding. We are not in favour of persistent benefit cheats, but the Lib-Dems are giving the impression that they are, by continually making excuses for them. We will not allow those excuses to go unchallenged.

Will the Minister of State give way on that point?

I will give way in a moment, and then the hon. Gentleman can make all the points that he wants. We have plenty of time for that.

It will not be the case that people will not understand what is happening. The people who we are talking about are not lovable rogues, but persistent benefit cheats. We could give many examples of people who have cheated the system twice or even tried to do so a third time. We catch and prosecute them, but all of a sudden, we still have to pay them benefit. My constituents, many of them low-paid taxpayers, wonder what the hell is going on. Why are we allowing that to occur? We have reached the point of saying, ``Sorry, but we are not doing this any more.''

Mr. Webb rose—

In a moment I will sit down, or give way to the hon. Gentleman, but I first want to make it absolutely clear that there is a set of benefits that will not count as a strike. They will not be withdrawn, even if a person defrauds the system and is prosecuted. They are maternity allowance, statutory maternity pay, statutory sick pay, working families tax credit and disabled persons tax credit.

Another list is of benefits that are disqualifying benefits but will not be sanctioned. One is the retirement pension. If a person is caught cheating on that, and prosecuted twice, it will not be sanctioned, but other benefits, such as income support, will be. Others benefits are disqualifying and sanctionable, including income support, jobseeker's allowance, and housing and council tax benefits. There are conditions for hardship allowance. For example, we will not withdraw free school meals, irrespective of the conditions. We have been very strict about what we will sanction. The hon. Gentleman mentioned proportion, but we are talking about the restriction of, say, £20.80 a week—or £10.40 for a person with children—for 13 weeks. That is not 26 weeks, or a year, or for ever. Housing benefit will remain, so the point about shelter is inaccurate.

Our policy is moderate in scope and tough in intent. It is designed to provide a deterrent and to ensure that the overwhelming majority of taxpayers and benefit claimants who are law abiding—neither benefit cheats nor persistent benefit cheats—do not suffer. The small minority who are persistent benefit cheats will be dealt with. The clause makes a technical adjustment of the benefit system to achieve that. I give way, as it looks like the hon. Gentleman has something more to get off his chest.

On a point of order, Mr. Maxton. The Minister has repeatedly refused to give way, so I would rather make a separate contribution.

I call the Minister.

I have finished speaking.

One can always tell when the Minister has things wrong—he will not accept interventions. He goes along creating a straw man in order to knock it down. The phrase ``lovable rogue'' came from his lips, not mine, and it was not said by my noble Friend Earl Russell. If the Minister had been listening, he would know that I said that everyone agrees that people who defraud the benefit system are doing wrong and deserve appropriate punishment. No one claims that these are nice or good people who should not be punished. A mature, grown-up debate would be about what punishment is appropriate. The Minister has simply asserted that the punishment in the clause is right on the basis, practically, of no evidence.

Let us examine the available evidence. Research reported in the Allbeson study suggests that a majority of people questioned did not fancy being ``close to destitution'' and might try harder next time. Any fool can impose a severe sanction to force people to do what he wants; the key question is whether, in a civilised society, the sanction is appropriate or excessive. That is our key point. It is not that we like these people or defend what they do; it is simply that the Government's sanction is not proportionate but excessive.

I specifically asked the Minister about children in affected families, but he ignored the question, perhaps accidentally. The report used the word destitution and the Department's researchers found that families with children were the worst affected. The Minister glossed over that, because it is good knockabout politics to pretend that other parties are soft on benefit fraud. What about the children? The Minister has not responded. One presumes that he believes that because parents should not defraud, it does not matter what happens to their children. These children suffer as a result of the present sanctions. The Minister is about to introduce new sanctions, which will mean more innocent people suffering. He offered no defence, because he has no defence.

Before we go further down the road of sanctions, the Liberal Democrats want the Department to recognise their effect—so far the Department has chosen to ignore what little evidence is available—and to take account of it in framing the pattern of sanctions. Instead, the Department has opted for this populist nonsense and suggested that we are soft on benefit fraud. We agree that fraud is wrong; we agree that it should be punished; but the question in a mature democracy is what punishment is appropriate. The punishment in the clause is inappropriate, so we shall oppose the clause.

Question put, That the clause stand part of the Bill:—

The Committee divided: Ayes 11, Noes 2.

Question accordingly agreed to.

Clause 7 ordered to stand part of the Bill.

Clauses 8 to 14 ordered to stand part of the Bill.