Previous Section Back to Table of Contents Lords Hansard Home Page


Lord Higgins: I am experiencing some difficulty in following what the noble Lord is saying. I believe that we agreed earlier that the word "likely" might well mean more than 50 per cent. However, the noble Lord is suggesting that it is likely that someone is contravening the rules, which is surely the same thing. To say that someone is likely to be doing it now is surely the same as doing it now. I am having difficulty understanding the distinction that the noble Lord is seeking to make by way of an alternative interpretation of the word "likely".

The noble Earl, Lord Russell, made a similar point about the problem with forecasting such matters. It seems to me that to say that someone intends to do something also poses a problem with regard to forecasting.

9 p.m.

Lord Grabiner: I respectfully agree with everything that the noble Lord, Lord Higgins, has just said. Apparently my noble friend Lady Hollis does not. We shall see how matters turn out. If the noble Lord, Lord Goodhart, is right, the words "or is likely to" add nothing to the words "is contravening". "Contravened" is the past tense; "is contravening" is the current tense; "is likely to contravene" is surely a futuristic prediction. Certainly that is how I read it. I believe that that is certainly the way that the noble Lord, Lord Higgins, reads it. That was what I tried to say when I spoke earlier.

I believe that "is likely to" contains the element of prophesying which is the point that the noble Earl made. However, I am not sure that the words "intends to" or "is intending to", or whatever the language is, are anything more than an exercise in prophesying for the future. However, as I say, at the end of the day it is a question of language that everyone is comfortable with.

Lord Goodhart: I hope that I may finish what I was in the middle of saying. The point is that if one uses "likely" in the sense of prophesying, then, more or less by definition, one will not be able to get any useful information because if the person is not doing it now but is likely to do it in the future, none of the information one can get from the various bodies will be of any use as they will not prove that he is committing any offence. The only point of getting information from those bodies is to give one evidence to support one's belief that the person is contravening a measure now.

Baroness Hollis of Heigham: There may be some inconsistency; for example, the two NI numbers that were mentioned at Second Reading which require an explanation. If the explanation is satisfactory, clearly

1 Feb 2001 : Column 891

the person is not likely to be committing an offence. But prima facie holding down two NI numbers suggests that there are reasonable grounds for believing that a person may be using those two numbers to commit an offence. We seek prevention rather than an after the event cure.

Lord Goodhart: In that case, rather than using the word "likely", is it not better to say that he "is intending to commit"? It may be a matter of better drafting. If a person has a present intention, that is a different matter. There may be evidence that a person is now doing something which is preparatory to the commission of an offence. But, if one is simply talking in terms of someone who is likely to commit an offence, then it seems to me one returns to the problem of looking at someone and saying, "He is likely to commit an offence in the future because he has committed an offence or several offences in the past". I could accept the word "intending", but it seems to me that as present drafted the reference to someone who "is likely to commit" an offence is deeply unsatisfactory.

Earl Russell: I hope that I may return to the points made by the noble Lord, Lord Higgins, and the noble Lord, Lord Grabiner. Both of them suggested that the intention is as much in relation to the future as the words "likely to". I recall an article by the late F W Maitland with the delightful title of The Early History of Malice Aforethought. Malice aforethought, mens rea, is a well recognised concept in the criminal law. The guilty intention is, if proved, a present fact. Here I think is the crucial distinction. If we can prove that there is a serious possibility that a person has an actual intention of committing a crime, we are proving that there is something in his behaviour--as my noble friend Lord Goodhart suggested, in his preparations or in his state of mind--which is evidence of an actual settled intention to commit a crime. That is something which the law is well used to investigating in court. Because it is, it has worked out a series of practical tests for doing that and it knows how to set about it. I am not aware that the law has any similar machinery for working out who is likely to do this, that and the other. When I think of the disagreement between my noble friend Lord Goodhart and the noble Lord, Lord Grabiner, both, I should have thought, people of great expertise in the construing of parliamentary statute, thinking of the case of Pepper v Hart, I wonder whether anyone reading our debates would be any the wiser about how to construe the words "likely to" and whether that is perhaps another reason for taking them out.

Baroness Hollis of Heigham: We have been having a lovely time. I hope that I shall not be seen as spoiling that. We are dealing with Amendments Nos. 20, 22, 29, 32 and 38. I shall start with Amendment No. 22 and then speak to the consequential amendments. The key change in the amendment is to expand on the term "particular person" by providing that that person must be identified by name or description. Behind that

1 Feb 2001 : Column 892

lies a concern expressed on Amendment No. 32 which is the link with the question of fishing and the need to identify more narrowly.

Where we know the real name of a suspect--that occurs in the majority of cases--we would identify them by name. However, there are also instances where a person lies about who he or she is. For example, organised gangs manufacture many different identities in order to claim as much benefit as possible in false names. In those cases we may need to identify suspects about whom we are inquiring by using other information. Generally, this would be the address where the person was living. I think that the use of the term "by name or description" which my noble friend has suggested admirably sets out our needs and I am therefore content to accept his amendment. I also hope that the Committee will bear that in mind when we discuss later the suggestion of the noble Lord, Lord Higgins, that we amend the term "particular" to "specified". The other changes the amendment makes seem to express the provisions differently but do not alter them in any substantive way. Amendments Nos. 20 and 38 are, of course, consequential to Amendment No. 22 and of course I am happy to accept those.

I now come to Amendment No. 32. There is clearly concern about the inclusion of the power to obtain information on people who are in groups which are more likely than others to commit fraud. I begin by explaining our thinking behind the provision. As the Committee knows, social security fraud is costing us over 2 billion a year, which is more than it costs to run the entire department. We have done a lot. We are measuring fraud in a systematic way. We have set targets. We have new checks. We have data matching. As a result, fraud is coming down. The latest figures show that fraud in income support and JSA has reduced from 9 per cent to 8.4 per cent. That is a tiny percentage but encompasses large numbers. Those figures were published in November of last year. The measurement exercises that we referred to at Second Reading--almost like a risk analysis--tell us a great deal about the type of person who commits fraud against the social security system. We know that fraud by people working and making claims costs over 300 million a year. We know that people lying about their family circumstances, usually because they are cohabiting but say that they are not, costs over 200 million a year. We know that income support claimants who have committed fraud previously are 70 per cent more likely--a statistic given by my noble friend--than others again to commit fraudulent income support claims. We know that claimants of income-related JSA are more likely than claimants of any other benefit to commit fraud. We know that lone parents who have recently separated from their partners are at greater risk than other lone parents of committing fraud.

Lord Higgins: Do we have any corresponding figures on working families' tax credit?

Baroness Hollis of Heigham: That is not a benefit; it is a tax credit. At present I have no evidence either way

1 Feb 2001 : Column 893

on fraudulent claims. What evidence I have seen suggests that there have been some concerns about childcare arrangements. It is in part based on the information sent by the employer back to the Inland Revenue. However, if the noble Lord wishes me to pursue the point on another occasion I am happy to do so. I return to the substantive point. I appreciate that the noble Lord cannot resist attempting to turn the WFTC into a benefit rather than a credit!

We need to make this information work for us. We want to use it to target our inquiries on those who are most likely to commit fraud--essentially, an analysis of risk. That is what lies behind the conclusion of the provision at new subsection (2C)(c)--the power to obtain information on a person who is in a group more likely than others to commit fraud.

The debate today has raised significant concerns, highlighted originally by my noble friend Lord Grabiner, about how this power could be used. There is significant concern that this power could be used to target people on criteria such as race, sex, disability, sexual orientation and so on--concerns which were raised by the noble Earl, Lord Russell, at Second Reading. I assured him then, as I do today, that that is not our intention; nor would it be possible under these powers because race, sex or sexual orientation, or the fact that an individual has red hair, wears glasses, or anything else, does not increase the propensity to commit fraud.

However, having said that, I do not make light of the Committee's concerns. It is true that the vast majority of lone parents are women. Although we would not select lone parenthood alone as a criterion for using these powers, I can see that the fact that most lone parents are women might cause concern. Listening to today's debate, I have wondered whether there is a different way which would allow us to make use of the information about the risk of fraud but which does not generate the kind of concerns raised. To that end I found the points raised by my noble friend Lord Grabiner and others extremely helpful. My noble friend knows a great deal about the problems we are facing and the need to have research information as the basis for targeting fraudulent activity.

Having listened to my noble friend's remarks, and those of other noble Lords, I am minded to consider removing the power to obtain information on those in groups more likely than others to commit fraud. I refer to new subsection (2C)(c). My noble friend believes that we can address our concerns in other ways. However, I should like to reflect further on the consequences of removing this power before coming back with a final decision because there are "read across" areas in other parts of the Bill. I refer, for example, to the use of reverse searching of telephone directories as a means of fighting fraud. I may need to make other more specific provisions to cover other types of inquiry.

1 Feb 2001 : Column 894

I take on board the Committee's concerns. I shall do my utmost to meet them by a government amendment on Report. Giving that undertaking in good faith, I hope that the noble Lord, Lord Higgins, will be able to withdraw Amendment No. 32.

Amendment No. 29 seeks to address three issues. I shall take each in turn. First, the noble Earl seeks to explore whether we should retain the words "is likely to" in paragraphs (a) and (b) of what will be new subsection (2C). We have had an elegant discussion about the futuristic prophesying nature of the phrase. I shall try to give an example of an instance where we obtain information which leads us to believe that a person is intending to commit fraud--he is about to commit fraud. We want to prevent that. Where we come across such information, we need to be able to act on it. For example, if we found forged identity documents at an address receiving benefit payments, we would need to conduct checks into the forged identities to establish whether any of them were being used to obtain benefit. Another example would be where a person was reported to be about to do something--to start work, or commence education. Clearly we would ask him about it. However, if he denied it and we had reasonable grounds for not believing him, we could check with the prospective employer or UCAS to verify his story. If we found that he was lying, we could stop his benefit from the time the job or course commenced.

The key point is when we come across information which leads us reasonably to believe that the person is more likely than not to commit fraud. That is why we have selected the words "is likely to" in this context. Having said that, I am attracted instead to the concept of injecting intent into the provision, as argued by the noble Earl, Lord Russell, and the noble Lord, Lord Goodhart. All the examples that I have given clearly indicate some idea of intent. However, again I should like to take time to reflect on the best way of capturing that concept in the provision. If the noble Earl will agree not to press his amendment, I shall undertake to return in good faith on Report with a new form of words that I hope will meet his concerns about intent.

The second aspect of the noble Earl's amendment explores the provision relating to groups, which I dealt with on Amendment No. 32. I shall not repeat my comments. I have said that I shall reflect on the issue and see whether I can come back to meet those concerns.

Finally, the amendment raises questions about information on family members. The noble Earl did not speak to that. Given that the issue may more properly be raised in a later group of amendments--his is almost a portfolio amendment--I shall come back to it later.

I have promised to reconsider the wording of the provision to inject the concept of intent rather than likelihood to commit fraud. I shall also consider removing the power to obtain information on people in groups that are more likely than others to commit fraud and I have accepted the drafting changes

1 Feb 2001 : Column 895

proposed by my noble friend Lord Grabiner, which would also narrow the provisions. In the light of all that, I hope that your Lordships might feel content.

9.15 p.m.

Earl Russell: I thank the Minister most warmly for everything that she has said. I entirely accept the need to reflect, probably at some leisure, on the drafting. The problems are complicated and with the best will in the world, which the Minister has offered us, they are not easily solved. Of course I accept the delay.

I am delighted by the Minister's comments about Amendment No. 32. I am also very much encouraged on Amendment No. 29. I hope that any draft on which we might hypothetically agree will cover all the examples that the Minister gave. That is not what we have been talking about. It is simply a matter of the wording and the concept. In hope and gratitude, I shall wait for Report.


Next Section Back to Table of Contents Lords Hansard Home Page