Public Accounts - Preventing fraud in contracted employment programmes - Minutes of EvidenceHC 103

Back to Report

HOUSE OF COMMONS

ORAL EVIDENCE

TAKEN BEFORE THE

PUBLIC ACCOUNTS COMMITTEE

FRAUD WITHIN WELFARE TO WORK PROVIDERS

TUESDAY 22 MAY 2012

ALAN CAVE and ROBERT DEVEREUX

Evidence heard in Public

Questions 1 - 172

Oral Evidence

Taken before the Committee of Public Accounts

on Tuesday 22 May 2012

Members present:

Margaret Hodge (Chair)

Mr Richard Bacon

Matthew Hancock

Chris Heaton-Harris

Meg Hillier

Mr Stewart Jackson

Fiona Mactaggart

Austin Mitchell

Nick Smith

________________

Amyas Morse, Comptroller and Auditor General, National Audit Office, Gabrielle Cohen, Assistant Auditor General, NAO, David Clarke, Director, NAO, and Marius Gallaher, Alternate Treasury Officer of Accounts, were in attendance.

REPORT BY THE COMPTROLLER AND AUDITOR GENERAL

Preventing fraud in contracted employment programmes (HC 90)

Examination of Witnesses

Witnesses: Robert Devereux, Permanent Secretary, Department for Work and Pensions, Alan Cave, Delivery Director, Department for Work and Pensions, gave evidence.

Meg Hillier: Chair, I need to declare that my husband is a non-executive director of an organisation that has a contract under the Work Programme.

Q1 Chair: Apologies to everybody who was kept outside for the private session. Can I just talk about rules of engagement, so that we are all clear where we are? I would appreciate, particularly given the time, the questions are answered directly and succinctly. I will interrupt only if I feel that that is not what is happening, or if we are going off what we are trying to establish. That is the first thing to say.

Let me also say absolutely clearly that this inquiry is not about challenging the Government’s policy around welfare to work; it is about satisfying ourselves as a Committee that you are providing value for money and that the way the Department administers the Government’s policies provides proper probity. You, Mr Devereux, as the accounting officer, clearly have an overwhelming duty both around value for money and as the guardian of the public purse. I think you would agree with that.

I sent you a whole lot of documents when we had our earlier inquiry into the Work Programme. We started with the Work Programme, although I have to say to you that this talks about contracted employment programmes, not the Work Programme -just to be absolutely clear what we are talking about this afternoon. I sent you a sample of e-mails and I sent you a copy of an A4e internal report, which the Committee also has and which we are publishing. In your response to me-your letter to me-what you said was that the allegations did not provide specific enough information to underpin an investigation. Can you explain to me what you meant by that? And do you mind speaking up because it is a very odd room, and earlier I could not get what you were saying?

Robert Devereux: As you know, I have professionally qualified investigators working for the Department. They deal with all manner of allegations and investigations. The first thing they need to do when confronted with an allegation is to decide whether there is enough information contained within it to proceed with a proper investigation. Sadly, in the cases that you sent to me, it was not clear how they were going to do that. That was their professional view.

Q2 Chair: Can I ask whether the people you put on to this work are qualified auditors and accountants?

Robert Devereux: Yes.

Q3 Chair: They all are. Can I ask why you did not choose to come back to me to seek further clarification if you needed it, rather than simply putting it to one side and saying, "Well, I’ve done this one. I don’t need to do anymore. There isn’t enough information here."?

Robert Devereux: Pretty much at the same time as you sent me those individual allegations, you will remember that the Department announced that it was doing a complete review of its contracts with A4e. So pretty much the same investigators went about doing a systematic sampled base inquiry of the contracts.

Q4 Chair: But in the A4e recruiter audit document, they were pretty specific. What you said to me in your letter was that it did not provide information specific enough to underpin an investigation. As I looked through that report, there are names of individuals, companies and recruiters.

Robert Devereux: But that’s not what I said-

Q5 Chair: And that’s why I cannot understand why you felt that the information was not specific enough and therefore you felt that you did not have to take it any further. You did not come back to me, except when I chased you on a letter.

Robert Devereux: With respect, I was going to write to you and I had a letter in draft when your letter came to me. I chose not to put that in the response because I thought you might think I was making it up, but it was true. You wrote to me about two sets of things and I have given you two sets of answers. In respect of the set of individual allegations-the individual e-mails-I have, indeed, said that there was not enough evidence there to do it. If you turn the page to the top paragraph of my letter about the internal audit reports, I have made it perfectly clear that there are cases there that we are indeed pursuing, so you are right that the internal audit report contains more information on which the investigators can write. That is not the case in respect of the many e-mails.

Q6 Chair: I was very careful to redact people’s individual details, but I did not realise that you believed that that did not give you enough information. I just do not understand why you did not come back to me to seek more information before saying, "There is nothing in this that I need to pursue."

Robert Devereux: Because I was in the process of going through a systematic review of A4e’s contracts consequent upon Chris Grayling’s announcement.

Q7 Chair: I understand that you did that, and it is a good thing that you did that, but I do not think that that is an answer to why you chose not to take, in my view, as seriously as I would have liked the allegations that I sent to you. I sent you them in redacted form to protect the individual and you simply said to me that the information was not specific enough. I don’t get it. I am trying to get a real answer out of that. I would have appreciated your coming back to me and saying, "I don’t think I’ve got enough. Have you got any more details? Could we contact the individual?" In some cases, I have permission from individuals, so you could have contacted them. I also would have thought that that might have helped you to uncover more of the systemic issues that the allegations suggest exist within A4e.

Robert Devereux: I am sorry that we did not come back earlier. I have come back. I have asked you for that information now.

Q8 Chair: You haven’t asked me for any information, actually. To be fair, Mr Devereux, all you have said is that I didn’t provide you with information that was specific enough to underpin an investigation-

Robert Devereux: I said, "If you have been provided with any further specific details relating to these cases, we will, of course, consider them further."

Q9 Chair: Well maybe it would have been better if your investigation officers had come and talked to me, rather than-

Robert Devereux: At the time, my investigation officers were doing a systemic check of what was going on at A4e. You do not get to systemic answers by looking at individual allegations; you do it by going through individual cases.

Q10 Chair: But part of doing a systemic check, Mr Devereux, is to take seriously individual allegations. I did not seek them out; they came to me unsolicited. I would have thought that if you were doing a systemic check, you would have taken them seriously.

Robert Devereux: Systemic means systematic, doesn’t it? Okay?

Q11 Chair: It means systematic, but part of the way in which you do a systematic check is that you look at allegations that are put around a particular company. I cannot understand why the hell you did not do that properly.

Robert Devereux: We clearly differ on the meaning of systematic; but, as I have asked you, if you have further information, please let me know.

Q12 Chair: Perhaps your officials will come and see me, so that we can go through the individual cases.

Can I ask you a second thing on the dossier that I sent you? In the recruiter audit one, there are allegations of nine instances of fraud, as defined by A4e. I would define it differently, but let’s accept the A4e definition. I think some of the reputation-to-company issues could well be defined as fraud. A4e only told you about three. What does that tell you about the company?

Robert Devereux: As I understand it, A4e’s view-but I am now checking it, which is why I am telling you in the paragraph at the top of the page that I am going through every single one of those nine-was that they had investigated it and there was no case. They thought they were telling me about the three that they had investigated and had concluded that there was a case to answer. All three related to one individual recruiter.

Q13 Chair: So A4e has now changed its mind on the nine cases that it had said are potential fraud in its own internal document. Is that right?

Robert Devereux: If you read the words carefully in that report, it says that there may be fraud in these cases, so A4e’s position-I am only repeating what they have told us-is that they did further work consequent on that review. They concluded that there was no fraud in six of the nine cases, so they told us about the three in which they concluded that there was fraud.

Q14 Chair: But at the time-

Matthew Hancock: What time was this?

Chair: 2008-09

Robert Devereux: That report was started on 16 June 2009.

Q15 Chair: In 2009, when they had a report that suggested, again, nine potential fraudulent cases, they only told you about three. If I can just say so, there are further cases, having read this document in detail, that they classify as reputational risk or where they talk about being unable to complete audit visit, but these are instances where, no doubt, Richard Bacon will want to say that employers did not exist or were just lap dancers or what have you. All those, to me, were potential fraud, yet you are quite happy to say only three in A4e. It tells you nothing about the company, does it?

Robert Devereux: I am sorry; let me just say it again. That report says they believe there could be fraud in nine cases. A4e, as a company, believe it did further investigation to conclude only three of them were fraudulent.

Q16 Chair: And you are quite happy-are you?-that the seven-

Robert Devereux: I am not quite happy, because my letter makes perfectly clear I am now reviewing every single one of those nine cases, and five other cases, which, like you, we conclude the Report gives us leave to do.

Q17 Chair: So you would agree that some are so-called reputations at risk-actually a third of the cases have a problem with them; some seven are reputational risk; some where they are unable to complete the audit visit-

Robert Devereux: So we are agreeing that we are going to look at those cases.

Q18 Chair: You are going to look at all the ones to date in that report-the 31% where issues were highlighted.

Robert Devereux: I have told you in my letter that we currently think there are 11 cases that we are going to investigate. If you would like me to tell you precisely which ones they are, I can do that.

Q19 Chair: Let me just ask you another one. In one case-again, in your letter-you say you are not taking this any further because the employer has ceased trading. Is that right? That is in your letter to me.

Robert Devereux: Yes.

Q20 Chair: Okay, can I just remind you that the Public Accounts Committee always does its work. We always examine past expenditure; that is the nature. In fact, one or two of our members occasionally accuse us of looking forward a little bit too much, but basically we focus on past expenditure. So I feel a bit surprised that there is a Devereux rule, which is the only way I can express it, that says that if you go into liquidation we are going to stop examining you. Is that in fact what you are saying to me?

Robert Devereux: Sorry. If the allegation is that company X falsified records and company X no longer exists-

Q21 Chair: I understand that, but you can still pursue the issue, unless you have the Devereux rule that says once a company goes into liquidation-

Mr Bacon: Can you just finish the sentence-after the word "liquidation"?

Robert Devereux: After "liquidation"-the issue is, what evidence can my investigators find?

Q22 Mr Bacon: Presumably, that which the administrator has, because when companies go into liquidation there is an administrator or a receiver. They do not just disappear in smoke or go in a shredder-well not always, anyway.

Robert Devereux: I do not have to hand exactly which dates we are talking about, so I do not know what would have happened to the paperwork.

Q23 Mr Bacon: Yes, but the Chair’s point surely is that the fact of a company’s having gone into liquidation is not a reason to stop investigating.

Robert Devereux: Exactly. It may be that my-

Q24 Chair: But that is what your letter to me says it is.

Robert Devereux: Okay. It may be that my prose is a bit spare. Let me check with the investigator whether or not-I don’t think this is a case that says, "If you stop trading, we don’t investigate you." I suspect this is shorthand for, "It stopped trading and we could not find any evidence," in the sense that I have just described.

Q25 Mr Bacon: So it did go in the shredder?

Robert Devereux: It might have done; I do not know. Let me check that for you.

Q26 Chair: Thank you for checking that for me. All I would say to you is that it is not acceptable to this Committee to say just because a company has gone into liquidation you failed to pursue an allegation of fraud.

Robert Devereux: But I have explained that I am not trying to say that to you.

Q27 Chair: Okay. The Committee would like to see the audit of A4e that you have completed. Can you let us have that please?

Robert Devereux: Sorry, which?

Chair: The Minister made a statement to the House. I have so many bits of paper; maybe somebody can find your letter to me. In your letter to me, you say that you have completed a-I will try and find the actual wording on it.

Fiona Mactaggart: The Minister’s statement says: "The Department’s Internal Audit and Investigations team undertook this audit,"-the audit that was announced in light of allegations-"supported by Ernst and Young, and examined the controls operated by A4E on all its current contracts with DWP."

Robert Devereux: And what my letter says is that we have provided, in the absolutely normal fashion, the completed version of the reports that we have done to the National Audit Office, which is standard practice. You may or may not have seen-but Ms Mactaggart will have-the answer that the Secretary of State gave to the parliamentary question of what our policy is on disclosing things to do with investigations. At some length-I apologise that it was in a three-page PQ answer-we tried to set out the thought processes that we go through when deciding what should and should not be published when it comes to internal investigations. There is a balance to be struck between the public interest in transparency and the public interest in the ability to prosecute successfully. I suggest that that answer of 30 April 2012 is the thing that-

Q28 Chair: So are you going to publish it? I don’t understand. I have not seen this stuff. Was this a question you asked?

Robert Devereux: It was a question from Ms Mactaggart.

Q29 Chair: We do not have that, so can you let us know whether you are going to publish it?

Robert Devereux: I will send it to you. I very much doubt that we will publish it, but we will have to wait for word.

Q30 Chair: Why?

Robert Devereux: Simply because, although I could show the National Audit Office, it is not self-evident that the things that turn up in an internal investigation report are sensible things to be put in the public domain.

Q31 Chair: The only thing I would say to you is that if you do not publish it-particularly in relation to A4e, which has excited a lot of public interest-you give an impression of apparent complacency, if I may put it like that. This is a whole lot of good chaps-I understand that the Chairman is an ex-Permanent Secretary, whom, no doubt, you have conversations with. It is a big company with an expensive PR machine, which we have seen in operation over the past few weeks. They are bound to be good citizens, so we don’t need to have open and transparent oversight of them.

Robert Devereux: Maybe I can read out the answer if you are going to go down that-that is not the reason.

Q32 Chair: No, I want an answer to that.

Robert Devereux: Shall I do that? "It is…not ordinarily appropriate to publicise the details of specific investigation cases… Disclosure of unproven…allegations could result in litigation… Disclosure of specific cases where the investigation identifies evidence of procedural non-compliance, but not evidence of prima facie fraud, could impact on the willingness of providers or whistleblowers to bring matters…to our attention… Premature disclosure of cases where there is evidence of prima facie fraud may potentially prejudice or damage future DWP or police investigations."

Q33 Mr Jackson: That may not be normal, but given that what we are looking at is, effectively, undermining the bona fides of the whole Work Programme, which is a very significant issue-not just about the contractual arrangements and efficacy of one company-I would respectfully suggest that disclosure would be helpful. I gather, Chair, that international law firm White and Case has concluded similarly with Ernst and Young that there is substantially a clean bill of health, but for the Department’s own good standing, as well as our remit as a value-for-money Committee, I think that should be looked at again.

Robert Devereux: I understand that. It is correct to say, as I said, that it is not ordinarily appropriate to publicise. Can I pick you up on what you have said? Nothing that has been sent to my Department gives me any evidence whatsoever that there is a problem with the Work Programme.

Q34 Chair: We are not just talking about the Work Programme, Mr Devereux, but we will come to that.

Robert Devereux: But your concerns are about value for money today.

Chair: But we look back.

Q35 Mr Bacon: The concerns are this Report, which is called "Preventing fraud in contracted employment programmes"-plural. Of course our concerns are about the present Work Programme, which we saw you about in February, but they are also quite rightly about other employment programmes, too, and that is what the NAO Report is about.

Robert Devereux: They are, indeed. Your colleague says that perhaps my judgment about how I do this is dependent on reassessing the bona fides of the Work Programme , but I am still waiting to see evidence that the Work Programme needs any bona fides established.

Chair: We will come to the Work Programme later in the session.

Matthew Hancock: On this point-

Chair: No, Matt. I have a queue of people.

Q36 Meg Hillier: Mr Devereux, the company A4e receives only British taxpayers’ money for its operations in this area. The law on freedom of information disclosure currently applies only to public bodies, but do you not think there is a need for a higher level of transparency in a company of this nature?

Robert Devereux: I do, and there is nothing in the answer you will see from the Secretary of State that tries to pretend that because it is at arm’s length, it is not covered by this. This issue is to do with the nature of audit work and whether such work must ordinarily be published. I have simply read out the answer. You asked whether I will publish it. The answer at the moment is that we do not ordinarily do so. We have not decided what we will do with this Report yet.

Q37 Meg Hillier: I think we have had the argy-bargy about the audit, but isn’t there a bigger issue about the types of organisations that the British Government-in this case through the DWP-have contracts with? If this was a shareholder-owned company, publicly quoted, there would be shareholders’ meetings and transparency. We have, in this case, with this particular company-I do not know about others across Government-a company with one major shareholder. We have received evidence and there has been a lot of stuff in the newspapers about its particular way of working. It does not have the same challenges as a publicly quoted company. Is it appropriate, do you think? Maybe you would be straying into political territory, but there seem to be real issues here about providing contracts to a company that is not transparent and accountable when it only accepts and spends taxpayers’ money.

Robert Devereux: That is a perfectly reasonable challenge. I think it matters whether or not evidence has been found that this company is fraudulent. To date, all the evidence that I have found is that it is not fraudulent. The issues to do with whether, in past contracts, there has been poor practice on their part and that of others is a separate question, because I am currently trying to establish whether I am at risk of fraudulent activity in my current contract.

Q38 Meg Hillier: Can we bring it up to date? Figure 11 on page 32 in part four of the Report gives examples of alleged improper practice in the Work Programme. It came from a National Audit Office examination of various correspondence received by the Committee or individual members of the Committee, which was passed to the NAO. It may include other sources, too. If we go to page 8 of the executive summary, paragraph 18 contains comments about how to improve things for the future. You have been saying that things were bad and are getting better or that we have not got evidence to prove the problems, but it states: "The Department does not systematically collect claimant feedback and complaints which could improve its awareness of the key areas where claimants consider they are getting a poor service." Paragraph 13, on page 8-it is just one of many examples, but I have picked this out-states: "Some of the key controls do not apply to some of the smaller programmes." It talks about payment being made before outcomes have been achieved. You can tell us whether that applies to the Work Programme. Altogether, that shows that there are real issues about how the Department is following through on information, and I wondered whether you wanted to comment on that.

Robert Devereux: This Report contains two quite different sections. Sections 1 to 3 are about whether or not taxpayers’ money is being paid on the wrong basis. There is another section, section 4, which is about whether the levels of service are as we would expect them to be, given the-

Q39 Meg Hillier: Actually, as someone who represents people in Hackney South and Shoreditch who go through the Work Programme and have gone through other employment programmes, satisfaction of the complainant and them getting a job seems to be absolutely linked to the value for money for taxpayers’ money, because, in the end, the taxpayer is funding it.

Robert Devereux: The simple distinction I am making is that this Government have chosen, having learnt the lessons of previous employment programmes, to nail its colours to well-defined outcomes and have deliberately not sought to define the nature of the service standard that an individual claimant should receive. In the New Deal programme, for example, we specified to within an inch of people’s life what process they should go through. We were not saying, "They will get a job." We were saying, "You will go through this process: 30 hours’ worth of work." All we are saying with the Work Programme is, "We will pay you if you get them into work and you will tell us, and you will publicise, what levels of standards you are prepared to offer to your claimants." If these companies do not succeed in managing their claimants well and ensuring that they get into work, they will not make money.

Q40 Chair: I know that you are anxious to get on to the Work Programme, but I want to get logic onto the inquiry. We need to look at what has happened in the past and then move to the future, because I think we all have questions about the future. I think that Meg was trying to ask about what it is in the record of a company that will make you decide-I am putting words in Meg’s mouth-whether they are proper for engaging with Government, whatever the contract. Forget about the Work Programme for the moment. How do you define "systemic"? How much are you going to have to look at before you decide that this is a company that it is not right that it should be receiving public money to provide a service to people? I think that is what she is saying.

Robert Devereux: At paragraph 2.6, I think, the National Audit Office has set out the four points that we sought to test bidders against when they bid for the most recent framework contract-that is the framework from which all new and future employment programmes will come from. With apologies, this is the current position not the past position-the current position is, in order to get onto the framework from which Work Programme providers and, indeed, mandatory work activity people were provided, they had to demonstrate that these four things were true.

Q41 Chair: And what about the record of the company?

Robert Devereux: In assessing this, we sought to put in place teams that actually knew about the interaction that was happening with the individual department-

Q42 Fiona Mactaggart: Would you be prepared to provide the Committee with evidence that A4e, for example, has in place a system for staff to report improper behaviour or performance management systems that avoid perverse incentives, because you heard the private evidence before us, which seemed to contradict both those things quite compellingly?

Robert Devereux: It did indeed, which is why, if I may repeat myself, you need to be clinically clear about which period-

Q43 Fiona Mactaggart: That is why I am asking if you would provide this Committee with the evidence upon which you depended to decide that it was a proper company.

Robert Devereux: Let me go and see. I cannot see what would be the problem with doing that.

Q44 Meg Hillier: You will correct me if I am wrong, Mr Cave-I may be wrong, because memory can deceive-but I think it was you at a previous hearing, or one of your colleagues anyway, who talked about not taking into account the previous track record of providers-

Mr Bacon: It was Mr Devereux.

Robert Devereux: It was me.

Chair: I think you backtracked on that in a letter to us afterwards.

Robert Devereux: And I subsequently explained that we had ended up at cross-purposes. There have been two procurement processes to choose people on the Work Programme. The first procurement process was to put people on to a framework from whence I could draw people down on the basis of the second contract. In the first framework thing, one of the conditions we set and tested was, "Have you performed on previous contracts whatever your business might be?" Because, if you remember, my point was that we were trying to bring in new entrants-so they will not all have done welfare to work, and one of the tests that we set was to ask if they had performed on previous contracts.

Q45 Meg Hillier: Performed well?

Robert Devereux: Performed well, exactly-this is a performance point that I am making. Some of the old employment contracts were not difficult to perform well against because all you had to do was to run a process, not to get people into work, so it was not very difficult for people to pass a performance test.

The separate point that I was making was that we were not limiting that to simply performance on employment programmes. So we did that for the framework and then, having got people on the framework, we have now selected people off it through a separate procurement for the Work Programme.

Q46 Chair: Of course, to get it clear, the Flexible New Deal was also outcome related-semi-but we have had an evolution over time. The Flexible New Deal was part up-front payment, but also outcome.

Robert Devereux: Correct. It got up to-

Chair: So it is not that the Work Programme is magically that different.

Robert Devereux: It is also 2009-

Q47 Austin Mitchell: You have just sat through a long closed session, which produced some fairly damning indictments of both the structures and the practices in A4e and Working Links and which gave several indications of possible fraud. Some evidence has been submitted to us as well-to the Chair-by the people who gave evidence in that closed session. Will those allegations be investigated?

Robert Devereux: They will be investigated if they are put in front of me. Since it was a closed session and you are alleging that there were damning allegations made, let me simply say for the record it wasn’t-

Q48 Austin Mitchell: They sounded pretty damning to me-I would like that on the record, too.

Robert Devereux: They did indeed sound damning. The thing that I have asked the Committee to consider carefully is to be really clear, when you consider the allegations, which programme they were talking about and which year they were talking about, because it is not my position to argue that past contracts had flaws in them. That is why we have improved, as the PAC would expect us to improve-

Q49 Mr Bacon: You mean, it is not your position to argue that they did not have flaws in them.

Robert Devereux: Did I miss a negative?

Mr Bacon: You did, but that is what you meant.

Q50 Austin Mitchell: Hang on, let me just follow this through. This is an area of difficulty because-I am talking about the new arrangements now-you have to check that the placements were in fact secured and that people went to them and stayed with them. There is a lot of checking there. You have to check whether they are part-time or full-time jobs. There is a lot of potential for argument and fraud. You have to take account of people who would have got jobs anyway, which is bound to be a substantial proportion of the people who go to these agencies, so it is an area of some difficulty.

Robert Devereux: I am afraid you’ve got that wrong.

Q51 Austin Mitchell: It is essential that checks be tight and that you have confidence in the structures of the companies doing the placements, and in your ability to check that they have actually been taken up. Consider what you have just heard, and the kind of evidence that has been in Private Eye from time to time.

Robert Devereux: A very reliable source.

Q52 Austin Mitchell: It reports that the records for work on the New Deal for Disabled People in 2007-08 were found on a desk, abandoned on a tip. They show that there were claims every month for sums between £400,000 and £1 million, for which you were routinely rejecting up to £20,000 a time. Without investigation, you were rejecting them when they were making duplicate claims and other sorts of mistaken claims. That is the nature of the companies that emerges.

Are you satisfied-after what you have heard of their control structures their audit systems, and their propensity to create a culture in which pressure pushes employees lower down the scale into fraud-that these are fit and proper companies to be doing the very important job, which I think it is right to do, under the new arrangements? Are you satisfied that those are fit and proper companies?

Robert Devereux: The most important piece of learning that my Department has taken out of their investigation of past contracts is that over-reliance on contractors’ internal systems is a mistake. As a consequence, we now have a system where it is very difficult to actually get the payment out of my Department on anything other than evidence that I personally can test. Let me repeat-it is in the Report, although I am not sure everyone has read it. For the Work Programme -I am sorry to be coming back to it, but this is the £5 billion programme that you are now worried about-the only way that you can get a payment from my Department now, when you submit the claim, is when, first, I can demonstrate that the individual you have claimed for has indeed been off benefit for the whole period that you have claimed for. Secondly, I then go in to check a sample of those against HMRC’s records on whether that individual was indeed in work. If they fail either of those two claims, I do not pay.

Q53 Austin Mitchell: That’s true, but you also have a vested interest in soft-touch investigation. The question was not whether you have new arrangements; it is whether you are satisfied that those companies are fit and proper.

Chair: Just answer yes or no.

Robert Devereux: I am answering this way because this is the right answer to the question. The question concerned what controls I have got in place. I have told you-

Q54 Chair: So is A4e fit and proper to undertake the Work Programme?

Robert Devereux: A4e is a contractor of ours, and it is subject to the controls that we have put in place, which makes me satisfied as the accounting officer that I am not wasting taxpayers’ money.

Chair: We’ll come back to that.

Q55 Matthew Hancock: Austin and I are brothers in arms on this. I was going to make the point and ask the question about accountability. It is incumbent on everybody to ensure that-especially when outcomes are more important-you can have certainty that those outcomes are being measured accurately, contrary to some allegations concerning previous systems. Can you take us through what you changed in the Work Programme to make sure that you have confidence over that, and which, therefore, it seems from your earlier answer, underpins your having less of a handle over the process that the providers need to follow?

Robert Devereux: Let me think which four I shall choose. I might choose the following four. We now have a much clearer definition of what the outcome is than we have had in previous contracts. It is fair to say, as the Chair said, that the Flexible New Deal was getting towards that. The truth is that the New Deal and the Work Programme have very well defined outcome definitions. So, for example, we are no longer working, in the case of the New Deal, on the expectation that the job might last 13 weeks, which is open to all kinds of abuse. Secondly, we have substantially reduced the number of contractors. When the New Deal was at its height, we had 1,300 contracts and 600 contractors; we are now down to 18 contractors and 40 contracts, which means that we can put in place a computer system that enables every single one of those transactions to be automatically sent to us by computer, not on great big paperchases.

Q56 Chair: When did this computer system start?

Robert Devereux: The computer system started for referrals in the summer of last year. As we went through with you in our evidence last time, the second piece started-

Chair: March?

Q57 Mr Bacon: Mr Cave, on that very point-this is completely germane-we had an exchange about this, and you said it would be in place by the end of March 2012. I take it that you succeeded in doing that.

Alan Cave: I took your wise counsel, Mr Bacon, not to send you a letter on 1 April saying that the IT system had been implemented. It was in fact implemented on 23 March, four days ahead of schedule.

Robert Devereux: I am going through my issues. So, first, there was a clear outcome. Secondly, there were far fewer people, so we could manage them directly. Thirdly, we have a computer system that is in there. Fourthly-this is the really critical piece-when somebody now applies for payment, it is no longer a question of a piece of paper coming in with or without some rubber stamp having been stamped on it, which somehow or other we pay and then have to chase and hope that their processes were good. When a referral comes in and a company says, "I would like to be paid for Mr Devereux. He has done this time off benefit and with this employer," Mr Devereux, in the case of the Permanent Secretary, gets the computer and says, "Okay. Can I just check with my system whether he is off benefit?"

Q58 Matthew Hancock: And that happens in each case, or is it a sample?

Robert Devereux: It happens in every single case before any payment is made.

Q59 Matthew Hancock: So would you argue that concerns about internal processes are less important because you have that outcome and, therefore, what would be the point of internal fraud within the Work Programme?

Robert Devereux: You might well ask, so let me finish the last check. The check that happens in 100% of cases before payment is made is the one that checks off benefit. Additionally-[Interruption.] I hope people are getting all of this, because this is quite important evidence. Additionally, we check a sample of all of those cases against HMRC records.

Q60 Matthew Hancock: Why not all?

Robert Devereux: Because there are a lot of them. We check them against HMRC records. If any of them fail the employment test-in other words, they don’t fail the HMRC ones-not only do we get recovery of the thing we have already by then paid, we also gross up that sample and take back the full amount. As an example, if 100 claims go through and I find that one of them is wrong, I don’t just get the money back on one of them, I get the money back on an appropriate proportion-

Q61 Matthew Hancock: So the incentive is to get it right?

Robert Devereux: Yes. On your critical point about internal processes, how is it now in anybody’s interest to say, "Let’s pretend this job exists?" If the job does not exist, it will not pass the employment test. What incentive is there for you to say, "This person was off benefit."? The only person who knows whether they are off benefit is me because I am paying them. There is no incentive in the system, so my interest in their internal systems comes to when it goes down, because I have taken the controls myself.

Q62 Matthew Hancock: On the question of the audit of past systems-you read out part of the written answer-it isn’t everybody’s view that that should certainly be published. I want to ask about your relationship with the NAO, because if you are going to show that to the NAO, we and the public need confidence that it will be scrutinised properly and be appropriately put into the public domain. How much detail do you share with the NAO? What does the NAO then do with it?

Robert Devereux: In my view, it is pretty common to show all your workings to your external auditor, if they ask for it. That is what they are there for. It does not follow, as night follows day, that if the external auditor knows, it must be in the public domain. So the answer to your question is that the complete audit report on mandatory work activity, which is the one we have completed, is already with Amyas. He has been talked through it. He asked me many questions, and I am pausing only on whether, having shown him, it must now be put in the public domain. That was the earlier conversation.

Q63 Chair: Amyas?

David Clarke: Shall I answer? So far, we have had three reports from the Department. The first one we received was on mandatory work activity, and yesterday evening, we received the other two of the three that they have completed. I can confirm they are the complete audit reports, plus one report about the fraud investigation into the mandatory work activity. Normally, those reports are not published. If the Committee wanted, we would do a memorandum that would summarise the key points, so that the Committee could have those.

Q64 Chair: I think some of us would rather have it published. May I come back to you on the Work Programme, because you make these assertions frequently, Mr Devereux? Let me just take you through another perspective on this. You have created a whole lot of new incentives in the Work Programme, so payment by results is the incentive, and that is how providers will make their money. It is a black box of activity that takes place, so you do not necessarily know how they are behaving.

Robert Devereux: No, sorry; that is not quite right. The procurement process did not specify the process they should go through. The procurement process asked them to say what minimum standards they were going to use, and I now regard that as contractually binding on them.

Q65 Chair: I understand that. So are you now demanding of them how they treat their clients as well?

Robert Devereux: Consistent with them in themselves.

Chair: Consistent with what?

Robert Devereux: Their published minimum standards.

Fiona Mactaggart: Their minimum standards.

Chair: Do you publish that?

Fiona Mactaggart: They are published on the website, and sometimes they are as vague as, "Keep in touch with clients."

Q66 Chair: Can I carry on, and we will come to them? As I understand it, a number of the companies-indeed, the two that we heard evidence from in private-award bonuses to their employees if they place people into work. That is true as well. It seems to me that that incentivises employees and companies-

Robert Devereux: I am sorry to interrupt again, but just to be really clear, I think you will find that in some of these cases they are paying team bonuses, not individual bonuses. That might make a difference to your sense of what this is.

Q67 Chair: Okay. Team bonuses or individual bonuses. Some are paying individual bonuses?

Robert Devereux: They are required-

Chair: I think A4e paid individual bonuses. I do not know whether they have stopped, but they certainly did.

Robert Devereux: For which contract-again, be very clear. They are required to make sure that the performance management systems must avoid perverse incentives-for example, to falsely claim outcomes.

Q68 Chair: They are required, but we will come on to the question of how you monitor that.

Robert Devereux: Through monthly contact management meetings.

Q69 Chair: What I am saying is that if there is a bonus system in place-if it is outcome-based-and if there is less concern with process, which you say is one of the strengths of the new system, I think those are all very good incentives, but it creates another set of incentives among providers, because they will only get their money if they get the outcome. They will only get their bonus if they produce the numbers. With the best of intentions by DWP, which undoubtedly was the case with your predecessors-if Leigh Lewis was sitting here talking about his Work Programmes, he would feel as passionately about them as you do today-there could well be an incentive to cut corners, to portray what they do in the best light and possibly to cheat or act fraudulently. Those new incentives exist.

Robert Devereux: Why would they be paid? They would only be paid if people are off benefit and in work.

Q70 Chair: Let me give you two examples, Mr Devereux, one of which I sent to you and you obviously did not think it was very important.

Robert Devereux: Is it a Work Programme example?

Q71 Chair: It is a Work Programme example, from 29 March 2012. This is a person who went to the Bexleyheath jobcentre-I have actually got it redacted here-but as soon as they were referred, they found a job: "I was just waiting for my new employer to open a dry cleaner’s, and I was offered employment. They said I still had to attend a course. I was suspicious of a job that was on their board"-blah, blah, blah. He/she got the job, and did not have to do anything. Then they wanted him or her to fill in a form. "I said, ‘I am not filling this in. It makes it look like you got me the job when I already had it.’ He said, ‘It is up to you, but you get a £50 voucher sent to you if you do.’ So I did, and I got the voucher in the post." The person went into the job, and A4e will get the money-they probably have not got it yet, but they will get the £3,000-for having placed the person in a job. But from the point of view of value for money, Mr Devereux, and our concern with value for money, A4e did absolutely nothing to support that individual into jobs. What I am saying to you is that the way you have constructed the new contract may have closed off some previous poor practice, but it has created new ones. That is one example. Let me give another example-

Robert Devereux: Can I deal with them one at a time, because I will not be able to remember them all simultaneously otherwise?

Q72 Chair: Okay. This man, woman, person got a £50 voucher, so that A4e could claim a job outcome.

Robert Devereux: Whose £50 was it?

Chair: A4e’s.

Robert Devereux: So I have paid A4e if this goes through-

Chair: For doing nothing?

Robert Devereux: For the rate we said in the contract for getting somebody-

Chair: No, they didn’t. They did nothing. This person got the job themselves, but they happened to have got the job just at the point-

Robert Devereux: Would you like to engage me then about what sort of test we are going to have for whether or not somebody helped somebody get into work? If you want to introduce a set of conditions that are impossible to police-

Chair: Our concern here, Mr Devereux, is not the rules you have established, but whether or not there is value for money for the taxpayer. This is one instance. I am about to read you another even worse instance. This is an instance where-[Interruption]-you can smile about this, but I think this is really serious. There is £5 billion of public money at stake here. The work provider is claiming for having done work without having done anything. So public money is going into a private provider company without any effort. I agree it has not broken your rules, but I said to you: do your new rules create new incentives. Let me give the other case.

Q73 Mr Jackson: Can I just ask before we go on to lots of anecdotes how you deal with that specific issue? What methodologies have you got in place to make sure that value for money is being looked at in a case like that? That may be a very atypical case.

Fiona Mactaggart: No, it isn’t.

Mr Jackson: Well, you have no evidence for that. What processes are in place to monitor the contract at that very low level?

Robert Devereux: I have to go back a bit to some of the evidence we gave you last time. Do you remember that a condition of these contracts and keeping these contracts is that the providers have to demonstrate that they have got 10% more people into work than would have been the case without them? So the bar to continue with these contracts is they have to beat what would have happened anyway. Part of my answer to you is that people are going to get into work. Some people will be easy to help and will make it back into work quite quickly. We have some relatively tight rules about not putting people on the programme who already have a job. That seems a rather obvious thing to do. There is an area that the Report talks about regarding people who suddenly find that they have got a job offer. Let me come back to the point I was making earlier. I am not sure that I could very easily contractualise the test "did you help me?"

Chair: No, you could process as well as outcome.

Q74 Mr Bacon: One of the things you could do if you are going to be spending thousands of millions of pounds on helping people into work might be helpful for you, as a Department of State, to have some way of assessing whether the contractor helped the potential employee into work or not. I do not want to put words into your mouth, but I think you just said that basically that was not possible. Is that right?

Robert Devereux: No, what I said was that if you are trying to construct a contract that you do not then end up-I come in front of this Committee and hear all kinds of questions about how does that process get adequately checked. We are at least now working on the basis that I have set the prices for this contract based on a detailed examination of 10 years’ worth of people moving in and out of work, which includes people who on referral pretty much get a job immediately.

Q75 Matthew Hancock: Hold on. So are you saying that you set the contract according to the average-the system-even if it means that some people will get out of unemployment the very next day? Obviously not everybody leaves unemployment the next day, otherwise we would not have a problem with unemployment. That would be a wonderful situation. I understood your answer to be essentially a defence of value for money at the level of the contract as opposed to a defence of value for money for every specific individual.

Robert Devereux: That’s correct.

Chair: We don’t know, Mr Devereux. With the greatest of respect, we are sitting around this table, and we don’t know. Can I raise my other case with you?

Robert Devereux: I haven’t answered the question yet.

Q76 Matthew Hancock: Can you answer my question? Is it your defence that it is the whole contract, and how do you therefore measure the macro and whether 10% is better than what would have happened otherwise? As an economist, I know that what would have happened otherwise is the most difficult thing to work out.

Robert Devereux: It is indeed. The story we talked you through last time was that in reaching the contract that we have in front of us now we did a very detailed analysis back over 10 years of HMRC data of everybody’s movement in and out of work-a one-day bit, a six-month bit, and a three-year bit-to try to establish the pattern of activity you could expect in a modern economy, against which we could then calculate what sort of price would be appropriate to pay that produces value for money. In addition, having established that baseline, we put in contractual conditions saying that unless you beat that do-nothing proposition by 10%, you are at risk of forfeiture, and we will ask for it back again-

Q77 Chair: Amyas wants to come in, and then I will do my other case and come back to you guys.

Amyas Morse: Just a comment. I wonder whether I might direct the Committee to another point. I think it is perfectly fair, and paragraph 2.14 of our Report says that if you have a large number of people with attachment payments, and being attached just before they start a job, it would show up statistically, so I don’t think it would be impossible for the Department to pick up on it. There is potential for that to be exploited, but it would be moderately visible data-wide. I don’t think it is too difficult for the Department to control. I am saying this against the clear view in the Report that the control environment has improved a great deal.

I don’t want to take away your right of reply, Robert, but I would say that what will be more challenging on a enduring basis will be making sure that when the contractors have published their minimum standards, within those there is real visibility when there are quite a lot of complaints from the public. The partnership in our view should have visibility of that and should require visibility of it, and make sure that those standards are pretty clear, and comply as far as possible.

Fiona Mactaggart: Yes.

Matthew Hancock: Exactly.

Amyas Morse: The reason for that is that we don’t want cherry-picking and so on. I am saying all this against a background that it is getting better, but none the less there are areas where more work could be done.

Robert Devereux: An observation is that "It’s getting better" was true of the entire document. The particular quote you said was that the Work Programme includes all the design, contractual, oversight and insurance controls that we set out as being best practice-[Interruption.] I will consistently come back and defend this, because actually, that is-

Q78 Chair: Will you answer the question, Mr Devereux? The question was about minimum standards, and whether you would be willing to publish the data around whether or not providers comply with minimum standards.

Robert Devereux: I am just trying to test what our practice is.

Alan Cave: Our practice is that every month in every contract our compliance officers examine in detail 25 cases, and take them back to individuals and say-

Q79 Chair: Will you be prepared to publish whether or not private providers comply with minimum standards? Can you just answer the question, and stop evading it, please?

Robert Devereux: We are trying to think before we answer. I think the answer is probably yes, but can I go away and think about it?

Amyas Morse: The question I was asking-sorry, and pardon me-was not just should you publish it, fine though that would be, but should you make a practice of gathering users’ complaints to make sure you get objective information. I don’t think it is too much to ask that you might consider that.

Robert Devereux: Okay. On a monthly basis, apart from the 25 cases, we have a routine standing agenda item to discuss the complaints that providers have received.

Amyas Morse: Do you get them yourself?

Robert Devereux: I don’t get them automatically myself. A standard practice, and you might like to ask the ombudsman what she thinks is the right way to do this, is for first-tier organisations to get on and fix first-tier problems, and then to have a second tier where things that have not been fixed go. To date, 18 cases out of 750,000 referrals to the Work Programme have got to second tier complaints. If that is the case, I don’t have a systemic problem with people worrying about levels of complaint.

Q80 Chair: Okay. The Report says that there is a low level of fraud.

Robert Devereux: It says a low level of what?

Chair: Of fraud. What is an acceptable level of fraud?

Robert Devereux: What do you think it is?

Chair: Well, you tell me. You’re the accounting officer.

Robert Devereux: Okay. My view is that reported fraud-this is a quote from the National Audit Office-is at 0.01%, right? So that’s one ten thousandth.

Q81 Chair: That is reported fraud. We heard in the session before that not necessarily all fraud is reported.

Robert Devereux: Yes, exactly. So maybe it’s a bit bigger than that. My guess is the level of fraud is going to be so small, you might wonder why this is the highlight of our-

Q82 Chair: What is acceptable? Your accounts are always qualified-I accept-for fraud on individual benefit, but fraud here, on the provider market, what is an acceptable level?

Robert Devereux: What is an acceptable level of fraud? The way Amyas looks at my accounts of benefit fraud, he has regarded the current level of fraud, at 2%, to be too high. We’re running a complex system and something in the percent-this Committee has never had a conversation with us and I should like to come back another day and have a conversation with you about-

Chair: This is preventing fraud.

Robert Devereux: About what would an acceptable level be.

Chair: You haven’t got an answer.

Q83 Mr Bacon: Before we move on, can I just clarify one thing? Paragraph 7 mentions 0.01% of spend. Does that mean 0.01% of spend on the Work Programme ?

Robert Devereux: No, on employment programmes.

Q84 Mr Bacon: In total?

Robert Devereux: It is £773,000 over £829 million.

Q85 Mr Bacon: So it’s all the other employment programmes, as well, that previously you were rather reluctant to talk about.

Robert Devereux: Well, there are-

Q86 Mr Bacon: Now you’re praying them in aid to make the number look smaller.

Robert Devereux: This calculation is £773,000 worth of cases where we’ve identified misrepresentation, against an annual spend of £829 million.

Q87 Mr Bacon: Across all programmes.

Robert Devereux: 82% of that £773,000 applied to the New Deal.

Q88 Matthew Hancock: How much applies to the Work Programme?

Robert Devereux: Zero.

Q89 Chair: Well, this is reported fraud.

Robert Devereux: There is no reported fraud on the Work Programme , either.

Q90 Chair: Can I ask about a case from Ann Coffey and how your systems would deal with it? I accept this is a case you haven’t seen-she just gave it to me this morning-of a man who started self-employment a week after he was referred to the provider-I don’t know which provider it is, sorry; this not an A4E case, okay, but another one-and they said, "Will you come and sign all these documents?", because they wanted to claim it as a job outcome. He was reluctant to do so. They then said, "Well, if you sign all these documents, you will get £2,000 after six months’ self-employment." So he signed all these things-signed and signed and signed-and at each meeting he was told that he would get the money. When he finally said, "Can I have the money?", of course it did not exist.

Again, I accept that this is an individual case, but I’ve said to you that I think you’ve created new incentives which could create fraudulent activity. How would your present new systems work with this sort of case?

Robert Devereux: First of all, I cannot see what the fraud against the taxpayer is in that case. I do sympathise with individuals-

Fiona Mactaggart: A self-employed person is a taxpayer.

Matthew Hancock: Exactly. He’s now self-employed and he was-

Chair: No, no, no. I think the bad value for money is that the provider did absolutely nothing for this chap, but claimed-I don’t know what you get-£3,000, or whatever it is, for him.

Robert Devereux: Let me make a point that we haven’t made in this conversation. I do understand the point about what happens if they do absolutely nothing. The contract here is designed both to get people into work and to keep them in work, all right? An awful lot of long-term unemployed cycle backwards and forwards, in and out of work. You get payments, as a Work Programme provider, for sustained work over the period of two years, so even if they don’t do anything to get him into work, actually, if they assist him in staying in work over a long period, rather than coming back, at the taxpayer’s expense, on benefit, that’s a good outcome.

So one of the reasons why you may think that we’re a bit too relaxed about whether they did anything to get him in is, actually, I want them to stay off benefit for a long time. I want them to be in employment for a long time. That is good for growth.

Q91 Mr Jackson: You’ve got processes to monitor periodically incrementally over that period, until he or she is considered to be a full-time, permanent person in work.

Robert Devereux: Not quite. We pay for most people after six months in work. We are then prepared to pay every month thereafter, up to a period of two years, if in that month they are in work. The incentive is to support people in staying there and not doing what too many people on previous programmes did, which is to go into work, collect the money and then fall back on benefits

Q92 Mr Jackson: So you are effectively saying that the cumulative benefit to the taxpayer justifies the fact that it might be a bit flaky at the beginning?

Robert Devereux: As I have already said in answer to your colleague, the aggregate value for money for the contract stands on the base of it.

Q93 Meg Hillier: A couple of quick-fire questions. Will you consider doing unannounced inspections to increase transparency?

Robert Devereux: Unannounced inspections are a perfectly legitimate thing to do if you are worried that, all of a sudden, investigated material has been thrown away.

Meg Hillier: It keeps people on their toes.

Robert Devereux: It does, but it is a function of investigation. This Report is talking about assurance teams. If I am sending somebody in to test whether their systems work, it is not self-evident that they need to arrive unannounced. The only issue that I take with this Report is that it is not distinguishing between two completely different things.

Q94 Mr Bacon: I am not sure that I follow that. This Report is about preventing fraud. You are not seriously suggesting that preventing fraud cannot be assisted by unannounced inspections, are you?

Robert Devereux: No, what I am saying is-

Mr Bacon: Then Ms Hillier is right, isn’t she?

Robert Devereux: No, what I am saying is that if I actually thought that the problem was such that I needed to turn up and knock the door down and quickly get in, then that is what I would do, but I also have a good reason to go back and check whether, for example all the things they have promised to do have happened. I might want to make sure that I turn up at the right place and they have the evidence together. I am not trying to make a big thing of this. Where it is appropriate, we will do that. Where it is not, we won’t.

Q95 Meg Hillier: That is very important. Would you consider having a single hotline for people who want to whistleblow or tell you about a problem so that they have somewhere to go? At the moment, people don’t think it is worth complaining.

Robert Devereux: I am less persuaded that that makes sense, for the following reason. At the moment, I do want people who are participating in the Work Programme to go to their provider if they have a complaint.

Meg Hillier: I hope they do.

Robert Devereux: If they don’t wish to go to their provider, they are seen fortnightly at their jobcentre.

Q96 Fiona Mactaggart: Do you audit what the providers do when people complain?

Mr Bacon: I just don’t understand your resistance to that. It is a very sensible suggestion.

Robert Devereux: The main reason is that people understand whistleblowing to mean all kinds of things. If actually what happensis they ring up and say, "I was supposed to have a meeting with my provider, but they didn’t turn up today," that is a complaint; it is not a fraudulent thing.

Q97 Matthew Hancock: As we have discovered ourselves this afternoon, it is important to have a process for whistleblowing, to ensure that when there are things that may on the surface look serious they are properly investigated.

Robert Devereux: Whistleblowing is typically a concept about staff whistleblowing on their organisation.

Q98 Fiona Mactaggart: But it is not just whistleblowing. The reason that I first wrote to the Comptroller and Auditor General about how A4e were behaving was that there was no process where the concerns of my constituents, who were referred to A4e, could properly be dealt with. I just got brush-off letters and there was no process where I could advocate on their behalf. Now that was not always about fraud, but there is a lack of a process.

Robert Devereux: The process is you complain to your provider. If you get no response to that, you are allowed to go to the independent case examiner, which all benefit recipients are allowed to go to. That is told to people when they go to the benefit centre. That is a second-tier complaints process. That is what the ombudsman votes for.

Q99 Meg Hillier: That is a bureaucratic answer, with all due respect. For an average claimant, it gets quite complicated.

Mr Bacon: What you really need is, "Phone this number."[Interruption.]

Chair: We are coming back.

Robert Devereux: I look forward to it.

Sitting suspended for Divisions in the House.

On resuming-

Q100 Chair: Can I just ask you about your new systems? Let’s get on to your new systems for assessing your new inspection regime. Again, let’s get the history right. You have had inspectors in the Department since 2009.

Robert Devereux: Do you mean investigators of allegations-that sort of investigator?

Chair: Investigators. They have been there since 2009.

Robert Devereux: No, I think in terms of 2009 you are talking about the construction of what we call the provider assurance teams. Those are Alan’s teams now.

Q101 Chair: Who are they? Have they got audit and accountancy qualifications?

Alan Cave: No, they are there to assure me that the systems and processes that the organisations are using are sound and compliant with what we said they would be.

Q102 Mr Bacon: Is this the financial appraisal monitoring team?

Alan Cave: It is what took over from that.

Q103 Mr Bacon: Admittedly, you have now replaced them. Presumably you then had some sympathy with the comment by one of our earlier witnesses that it was totally ineffective as it was.

Alan Cave: The two things go together. It is not that the financial appraisal monitoring teams were, in themselves, ineffective, but they were doing the wrong thing.

Q104 Mr Bacon: Quite. One of the questions that I have never understood, is why they didn’t look at the internal audit reports. The answer that the Report gives is that you say it was because the information that you might get might vary from one part of the country to another. That sounds like a statement of words, but it is not really an answer to the question.

Alan Cave: I was making a separate point. The reason they are doing the wrong thing is because at that stage, the controls environment was all about pieces of paper, carbon paper, claims for jobs being expected and bits of paper back from employers. The FAM teams were there to do a sort of tick-list sampling that those forms had been filled in correctly.

Q105 Mr Bacon: But if you wanted to assess the quality of the control processes, looking at the internal audit report is bound to be informative and helpful, isn’t it?

Alan Cave: I agree. I’ll come to that point-

Mr Bacon: Hang on. Mr Cave is nodding and saying, "I agree", and Mr Devereux is shaking his head.

Robert Devereux: I said we agree.

Mr Bacon: You were shaking your head while agreeing?

Robert Devereux: Amazing, isn’t it?

Q106 Chair: Explain to me. If you agree, why in 2011 when your team visited A4e, did you not ask for the internal audit reports?

Alan Cave: In 2011? I am not sure that is the case actually.

Chair: In 2011, paragraphs 33.22 to-

Alan Cave: So this is when the PAT team did it. Can I make a point about the internal audit? Personally, to put my hand up on this, I think we missed that 2009 internal audit report and that was wrong. I should have been on the look-out for that because it was referred to in the Work and Pensions Committee, and I should have had somebody covering that. As we have discussed, I do not think anything in that report materially changes anything we knew and were on the case for, but we should have got that. There is some good counsel in the NAO Report that we should more systematically make sure that we understand what the audit plan is for each of the-

Q107 Chair: That is a different thing.

Alan Cave: No, it’s not-excuse me. Can I finish my sentence?

Chair: Actually this bit of the Report says that in 2011 you didn’t ask for it.

Alan Cave: And I am acknowledging that that was probably something we should have done.

Chair: No, you said ’09.

Mr Bacon: You did say that was a fair cop, I think.

Alan Cave: I was talking about the 2009 one, and saying that I think the NAO makes a fair point that as a matter of routine we should, every time we start a PAT investigation-as we are doing now-check through the audit plan and see which of the internal audits look relevant to the competency of the PAT team. I think that is a fair cop.

Q108 Mr Bacon: This high level of assurance is obviously going to cost a higher sum of money. Can you say how much more, in particular but not only in this? We took evidence on the Work Programme Report at the beginning of February. It refers to the computer system-the management information system-as originally budgeted at £4.6 million, and having underestimated the budget now stands at £8.6 million.

Robert Devereux: He is talking about the previous Report.

Alan Cave: Okay.

Q109 Mr Bacon: Yes, I was making that very clear. I know, because we had an exchange about it and you told me it started work on 23 March, so I’m hoping you will be able to tell me how much it cost. Was £8.6 million the final number?

Alan Cave: It was a little bit below that, I think.

Mr Bacon: What is the total cost of this higher level of assurance compared on a like-for-like basis, if you can, with the rather looser regime before?

Alan Cave: That is an interesting question. Because we have moved to a regime that now depends on off-benefit checks, computerised processes and us taking on board a lot more of the automatic checking, fewer people are needed in the PAT structure than when it was the old FAM team. Under the FAM team, we had scores of people chasing around looking at bits of paper.

Q110 Mr Bacon: Because you are worried less about what goes on inside the provider, than about your checks with HMRC and so on.

Alan Cave: Exactly.

Q111 Mr Bacon: I was talking to the Minister about this in the Lobby-about the extent to which and how one might go about defrauding-and he said you had appointed UBS, with the task of asking those sorts of question. With the projects now, one of our earlier witnesses said that if you dangle the carrot, the donkey will move. Well you’ve still got a carrot-there is lots of money. If you were a donkey, how would you go about moving? It seems to me that since the thing requires not only an up-front payment to be validated, but a sustained payment later when the job has been maintained, you need to convey the convincing impression of a job that is created, or secured, through the programme and sustained. That would be your starting point.

Alan Cave: And off-benefit.

Q112 Mr Bacon: Yes. Obviously, you’ve checked that: you’re not paying any benefit, so that is a saving from your point of view-if you were paying benefit before-but you still have the issue of whether a company, a provider, is getting those two payments for nothing more, potentially, than having created the convincing impression of a job that is sustained for two years. Every system is open to gaming and that would surely be how they went about it.

Alan Cave: How would you do that?

Q113 Mr Bacon: I am not about to go into too much detail in case it gives you too many ideas. Presumably if you are relatively lowly paid you would not come to the attention of HMRC because we’re shoving up the tax threshold, so a check with HMRC would reveal nothing.

Alan Cave: No, it would, because it is a P45 check; it is not a payroll check.

Q114 Mr Bacon: On the current HMRC website it is possible to create your own P45. I was told this by a PAYE journalist recently. You can just put in what you like. Whether they have pulled it down now, I do not know, but it was up there very recently. The fact is that if you are lowly paid, you would not necessarily be paying tax-you might be working black-and the provider has got its money, so they are both happy.

Robert Devereux: But if you are working black and there is no record in HMRC, you will not get paid.

Q115 Mr Bacon: Sorry; you mean that the provider will not get paid?

Robert Devereux: Yes.

Q116 Mr Bacon: Even if the tax paid by the individual was zero, there would still need to be a record at HMRC with a national insurance number, before the person got paid. Is that right?

Robert Devereux: Yes. Let me elaborate to be clear. We make an automatic check in this sample with HMRC records. Just because it is possible, hypothetically-HMRC records are not wholly correct-if it fails that test, we will go and seek to find whether there is other evidence of employment.

Q117 Mr Bacon: It is said there are about 75 million to 80 million national insurance numbers out there, which is significantly more than the number of people. I have not investigated the secondary market for national insurance numbers, but presumably it exists. How would you guard against that possibility?

Robert Devereux: There are a lot of dead people for whom we still need to know the national insurance numbers, because their spouse’s pension depends on it. That is one reason why there is a large number of NI numbers.

Q118 Mr Bacon: But there is fraud as well.

Robert Devereux: There is fraud, yes. I have not come to tell you that the system we have put in place is absolutely foolproof.

Q119 Mr Bacon: No, and I do not expect that.

Robert Devereux: I have come to tell you that it is orders of magnitude better than it was previously, and I am hoping you are going to be pleased with that. On that basis, we are checking, with HMRC-I cannot think who would be better to check, because it is supposed to know if they are in employment-and if the HMRC record is found, we will go and find an employer. I agree that it might be possible to construct some circumstances to get past all of that, but I am struggling to think of it, compared with-

Q120 Mr Bacon: I appreciate that you have tried very hard. I think what you have is a clear improvement, which the NAO Report says. Your direction of travel is terrific, and when we look at this again in two years, I am sure there will be a nice graph showing how successful it all was and how much benefit expenditure has gone down.

Robert Devereux: I look forward to coming back then.

Q121 Chair: If you are still there.

Robert Devereux: I will be.

Q122 Mr Bacon: Marvellous. We will hold you to that as well.

One more question: are providers contractually obliged to tell you if the beneficial ownership of the provider changes, as is quite common in Government contracts? When Emma Harrison resigned as chairman, she was reported as being-in fact, Mr Dutton told us-the 87% shareholder, so we can assume, in the absence of evidence to the contrary, that she continues to be the 87% shareholder. Is that right?

Alan Cave: Yes, because if they have a change of ownership, it is a contractual condition that we need to agree that.

Q123 Mr Bacon: Who are the other four owners?

Alan Cave: I do not know.

Q124 Mr Bacon: You do not know. Is that not a question you would be interested in? What if they turn out to be people whom the Home Secretary judged were not conducive to the public good and should not be in this country? It is not a trivial question. I am asking it in all seriousness.

Alan Cave: I am not treating it as a trivial question; I am just trying to reflect whether we would have known it at the time of the bid. I can find out and let you know.

Q125 Mr Bacon: Could you send us a note on that?

Alan Cave: Yes.

Q126 Mr Bacon: I would like to know who they are. I learned only the other day that Sir Robin Young is the chairman of A4e. Is that still correct?

Alan Cave: I believe it is, yes.

Q127 Mr Bacon: Do you know if he is a shareholder?

Alan Cave: I do not know.

Q128 Mr Bacon: Mr Devereux, have you had any conversations with Sir Robin about this at all at any point?

Robert Devereux: None whatsoever.

Mr Bacon: None whatsoever. Thank you.

Q129 Chair: Ever since you have been in post?

Robert Devereux: Ever since I have been in post. I thought you might ask me that question.

Q130 Chair: It is just that I know the old boy network.

Robert Devereux: Do you? Well, you do not know this old boy.

Mr Bacon: The girls are excluded. Didn’t you know that? It is a first principle.

Chair: The girls are excluded.

Q131 Fiona Mactaggart: One of the things that I am struck by is that, in order to improve the ways of reducing fraud, you are depending on one kind of control, which is a financial control. In effect, you are choosing not to use other kinds of control which many companies would use-transparency about the process within the company, whistleblowing records and so on. These things are not actually part of your system. I am wondering why you have chosen not to make them part of your system. It seems to me that that is a deliberate choice. On the financial outcomes, your making that system more robust than previously might deal with some past frauds, but these other things would deal with another kind of fraud, which is the risk of really bad quality provision-risks other than just straightforward financial fraud. I am wondering why you chose not to bother with that.

Robert Devereux: There are two answers. First, we have not chosen not to bother with it, because we have conversations with them month in, month out about what their processes are and what they are doing. It comes back a little to the nature of what we are trying to achieve here. What we are trying to achieve here is to get people who have been on benefit for a long time into work, and that is a thing that I prize above everything else. I am trying not to specify what sort of process they all go through-quite deliberately. I am not trying to be silly with this, but you cannot defraud the DWP by some other process if the controls are as good as I have just described.

Q132 Fiona Mactaggart: I understand that, but for example if you tested schools only on their examination results, the schools could put in fraudulent people to do the exams. I am not suggesting that that happens, but I am saying that you can substitute one kind of fraud for another. I am concerned that by having merely a one-legged stool, you put yourself at risk of other kinds of fraud.

Robert Devereux: It depends what your analogy is. My view is that I have very tight control over the cheque book-that would be my analogy. The people who are coming to the programme have been referred to them by me. I know their name. I know their national insurance number. They were a client of mine. I am not going to invent a pupil, because I sent them there. When they come back and say, "Mr Devereux’s got a job", if Mr Devereux wasn’t referred, I might smell a rat. Quite seriously, I am struggling to think what the obvious fraud is-particularly the sort of stuff you have heard this afternoon-that would be perpetuated against me. I absolutely take the point that maybe there is a very sophisticated construction-

Q133 Mr Bacon: Just take the diesel washing case in Northern Ireland, where HMRC said, "This is absolute nonsense. It would be physically impossible to do diesel washing to get rid of the red diesel on the scale required, so it is not something that we’ll take seriously as a big risk." About a week or two later there was a story in The Times about a diesel washing operation on a scale that simply staggered all the HMRC officials.

Robert Devereux: So let’s not be complacent. I agree with that. There might be things, and we need to keep alert, right?

Q134 Mr Bacon: Is your mindset therefore that you have eyes in the back of your head, looking around and scanning the whole time?

Alan Cave: It is, and one of the things that we need to do on the back of the NAO Report-it gives us some good pointers, actually-is to think about how our various control mechanisms, in terms of teams, are configured. They all do a very good job. I think that they are excellent, actually. The current configuration is designed for the transition process, which we have been through, to get from old world to new world. They have done that very well. There is a case, and you make it implicitly, that perhaps now is the time to have a bit of a look to see, with an eye to the future and looking round as many corners as possible, whether we need to tweak or look at some of the structures.

Q135 Chair: Is that to look at this thing that we-well, I-have been banging on about about bringing in people, auditors or accountants, who know a little bit about money?

Alan Cave: We are not short of auditors or accountants.

Q136 Chair: You said that you did not have them. You said that you had them in the team that went into look at A4e, but do you have them in your performance whatever-you-call-it?

Robert Devereux: In the provider assurance teams, we do not routinely ask for auditors or accountants. One of the recommendations in this is that we should think about the skills, and we agree.

Chair: You agree.

Robert Devereux: Yes, we need the team to be adequate. I do not need everybody to do it. These are helpful suggestions that will help us to be-

Mr Bacon: If you are not careful, we’re going to sing the departmental song together.

Q137 Fiona Mactaggart: One thing that this Government have said, and there is a serious point in this, is that increased transparency can mean that the citizen plays a role in auditing the quality of a service, if they can see what is going on. I accept that the problem here is that we often have a reluctant customer, namely, the workless person. What struck me from my contact with people is that a large proportion of those customers are very often not reluctant, but are very keen on getting into work, very keen that the public’s money is properly spent and feel that what they see is not spending the public’s money properly.

I looked at the Work and Pensions Select Committee Report on the old system and it said, "the ‘black box’ applies only to how customers are helped by providers; it never applied to auditing or to the treatment of subcontractors. The Department has also specified in the Commissioning Strategy"-I noticed that the commissioning strategy which was extant then, is still on your website-"that its assurance processes would ensure that the hardest to help were offered a good service by providers."

One thing that I keep saying on this Committee is that value for money is people getting good value, as well as the taxpayers’ pound being spent frugally. I am concerned that as well as fraud, we have some examples of very poor value, "I’ll give you 50 quid"-or £2,000 even-"if you fill in these forms". The new systems runs the risk of such gaming.

Alan Cave: There is more transparency in this programme and these contracts than in any of its predecessors. All the terms of the contracts are out there on our website-never had that before. You may have a criticism about the minimum service levels, but they are out there.

Q138 Fiona Mactaggart: Wouldn’t it be a good idea to have a minimum service standard that was universal?

Alan Cave: No.

Q139 Fiona Mactaggart: Why not?

Robert Devereux: Who would define it and why would it be right?

Q140 Fiona Mactaggart: Why don’t you look at the minimum service standards, which are very diverse, between the different companies? Actually, in some ways, A4e’s is better than many, because it is more specific.

Alan Cave: What conclusions do you draw from that?

Q141 Fiona Mactaggart: One of the things that I think is that you should establish some minimum services standards that are perhaps universal. I understand why you did not do that at the beginning of the programme-fair enough. I think that giving people the chance to go for minimum service standards at the beginning of a programme can create more competition.

Alan Cave: The conversation that I really don’t want to have in two years’ time is one with a provider who says, "My performance would have been so much better if you hadn’t insisted that I put in those minimum standards." That would have sent us back so far. It would be a terrible, retrograde step.

Q142 Fiona Mactaggart: So you think it is fine that some contractors have minimum standards that mean that they can be a three-hour journey away from someone who hasn’t got a car.

Alan Cave: If that was what they bid for and if that is what they think is consistent with getting more people into work and keeping them there, yes.

Robert Devereux: At the end of the day, we have thought very hard about this. In figure 12, there are examples of what individual providers have actually chosen to do. Even in respect of something called, "Meeting with personal advisers", even those who had specified a minimum standard cannot agree on what that should be. I do not see why you would imagine that Alan Cave’s judgment about the minimum standard, as a necessarily good thing to do-

Q143 Matthew Hancock: Amen to that. On the current evidence before us, you are a very impressive performer. Have you found that the fact that lots of different providers have looked at different ways of doing it has thrown up new ways of working to improve the ability of providers to get people into work?

Alan Cave: That is a really good question. I think that is beginning to happen. The key to a successful Work Programme is that it uses the flexibilities in the contractual arrangements, plus the much more real-time performance information that it now has, because it is there and can understand which interventions work better for which groups in getting them back to work faster. It can use that combination to innovate, change and flex. Cutting across that with a state-imposed set of minimum processes is antithetical to that dynamic, which I think is how you get improvement.

Q144 Amyas Morse: I have some sympathy with what you are saying, but that is different. If I can just complement what you were saying, Fiona, what about saying that we are going to keep on expecting them to say how they are going to improve? As they put their hand up and as they paid for the contracts, they did not know what it was going to be like when they put their bids in. Rather than saying, "Let’s have one size fits all"-I understand what is wrong with that-surely there is nothing wrong with you saying, "We will want to talk to you about how you are going to improve these standards of service over a steady period of time so that you are clear and more focused, it is clear what you are saying, and people can better understand the deal." Unless you think you have it all right first time, I would have thought that that made sense.

Robert Devereux: We don’t think that, but remember the structure of this contract. At the moment, we are paying people to be attached to the programme. That is not a job outcome. That payment is going down in each of the next three years and will eventually be zero, so half of that will just be- They are going to have to get better just to keep their cash flow going.

Q145 Fiona Mactaggart: Or they could go bankrupt if they’ve taken £8.6 million out of a company and run off with it.

Robert Devereux: They could go bankrupt and the consequence of bankruptcy would be that I would not have paid, with taxpayers’ money, for outcomes I didn’t get.

Q146 Fiona Mactaggart: I understand that you wouldn’t, but she has already pocketed £8.5 million.

Robert Devereux: To date, the evidence from most of the employment programmes prior to the Work Programme is that they have not actually achieved very much, so we need to achieve more; otherwise, we will continue with long-term unemployment.

Amyas Morse: I understand that and am really not disagreeing with it. I am just saying that there is nothing wrong, given that you are meeting them so often, with having a discussion and saying to them, "There are elements in this that we would like to see clearer and better articulated. Let’s talk about it."

Alan Cave: That is fine. The only rider I would put on it is "as long as that is a discussion linked to performance", because I want to know what they are doing to improve service levels such as to get more people back to work.

Amyas Morse: I agree with you.

Q147 Fiona Mactaggart: One discussion that could link to performance is that you could require every supplier to have, targeted at their staff, a whistleblowing procedure that was publicly available.

Alan Cave: They have-that is a requirement.

Q148 Fiona Mactaggart: It is a requirement. Are you confident that the staff of these companies know that, because it is not what I am hearing from people who work there?

Alan Cave: That is something that the PAT teams, as they go through all the Work Programme providers, will investigate and look at, and if they find they are not getting-

Q149 Fiona Mactaggart: Yes, on pre-announced investigations, so the posters are up nice and noticeably while you are there, Mr Cave.

We keep hearing from people that they do not know how they can complain and that they have bullying managers who prevent complaints. I accept that you have not been able to investigate all these cases, but one of the things that I have learned as an MP is that while you get green ink complainers, when you get the same complaint from all ends of the constituency, there is something that you should be alert to, and one of the things that I am very struck by is the number of people who have worked for these companies and say that bullying management means that there isn’t a way in which they feel that they can safely make complaints when they are asked to do things. We have had evidence from people who have been asked to do things that include making up evidence and telling lies. We have not been able to test that evidence in a court of law. Nevertheless, it is quite perturbing that people feel that that is what they have been asked to do and that there is no way they can whistleblow about it.

Robert Devereux: Let me come back to the rather important point I made earlier. In the evidence you heard, we were swooshing around the years. It has been the case that some of the things you have heard-

Q150 Fiona Mactaggart: I am talking about evidence that we have heard about the period of the Work Programme of someone being asked to lie.

Robert Devereux: In respect of the Work Programme?

Q151 Fiona Mactaggart: It wasn’t in respect of the Work Programme. You’re quite right-you were sitting here. She was talking about a different programme, but it was in this period, and therefore she did not know how to complain. While you are right that the programme that she was asked to lie about was not the Work Programme, if she had known of a safe way of passing it up, that would have equally applied to people who worked on other programmes at that time in that company.

Robert Devereux: I am going to check something and then drop you a line, because I think some of these providers have actually established independent confidential helplines. In other words, they are not whistleblowing to staff; they are whistleblowing outside the organisation. I’d like to check that before I quote it.

Q152 Mr Bacon: Some of them.

Robert Devereux: Well, I have evidence of a provider who I think is doing this, but before I-

Q153 Fiona Mactaggart: Can you tell us how many uses of these so-called independent lines there have been?

Robert Devereux: If we knew, I would happily tell you.

Q154 Mr Bacon: Could you send us a note and in it state how many providers there are in total under the Work Programme?

Alan Cave: Eighteen.

Mr Bacon: It’s 18 private contracts, isn’t it. [Interruption.] Of those, how many have established this independent confidential helpline and how many haven’t?

Chair: I think we’re coming to an end. We’ll hear from Austin now, and then I have a couple to sweep in at the end.

Q155 Austin Mitchell: I’m not sure the Committee wanted to bring me on, but I think Mr Devereux has said that one of these companies could go bankrupt. That is about equivalent to the Governor of the Bank of England saying moral hazard could discipline the banks because they might go bankrupt, too. They are already too big to fail. It is so unlikely. Here is a new category of capitalism that was set up to milk the state. It is not facing the competition of the market; it’s got a safe contract, which just goes on, so the threat of bankruptcy is really hypothetical.

Just let me ask this, though. Let’s not take A4e; let’s take the other one-Working Links. A number of disreputable things have happened with Working Links. A lot more was revealed when the papers for the disabled work scheme were dumped on a tip for public exposition. It was a curious filing system, but it did bring things out into the open that we didn’t know about, and you’d had compliance visits that had proved fraud in several of the local organisations. So, here is a case against Working Links, which is obviously badly organised and badly run, yet at the end of the day, you’re getting this new scheme-I think the new scheme is a good idea, actually. Payment by results is a good way of doing it-it’s very sensible-but it tends to be inferred that getting people a job drives out all other considerations when you say that’s the prime objective. Why was a company with that record and that structure reappointed to the new system-it is now running the system in Wales?

Robert Devereux: In the closed session I heard evidence that dated back to 2008 and 2007-that was what the gentleman sitting in my seat seemed to be saying. The process that we talked you through, by which people got on to the Work Programme, was one in which at the point that we were doing it, we checked whether we had concerns about their fraud. I have deliberately said several times now that the process pre-2009 frequently was problematic, partly because it was a paper chase with dodgy definitions, but also because I don’t think that there was a systemic segregation of duties, so if I am the person who puts you into work and I am the person who makes the claims to DWP, there are all kinds of incentives and opportunities for me to do that. If you introduce segregation of duty, that becomes a different thing. That was what the colleague who spoke earlier was talking about, because most of his evidence of fraud was actually things that had been tracked in the company. My view is that these companies have been on a journey and I am not in any way going to condone poor behaviour in a period when they had the wrong systems.

Amyas Morse: You have just taken away the contract for mandatory work activity.

Robert Devereux: I have, but not for fraud. I am trying to be clinical in my answers.

Q156 Austin Mitchell: Was there a lot of competition to take up the contracts?

Alan Cave: There was a lot of competition.

Q157 Chair: Mr Devereux, every permanent secretary arrives and says, "It was all rubbish before me, and it’s all going to be brilliant since I have been here." You are pretty normal in doing that. I have tried to say this before, but from our point of view, we look at past expenditure. I try to haul back perm secs who may have moved round government to call them to account so that they become accountable as well as being responsible. It is a bit more difficult in the case of Leigh Lewis, whose record you are attacking, because he has left the civil service, but all of us who knew him think that he was a very upright, passionately committed individual.

On A4e, Mr Devereux, you talk about the allegations going back to ’08 and ’09. Some do, but some of them relate to what has happened since, on your watch, in 2011-12-I don’t know whether you had joined the DWP then, so perhaps they did not happen on your watch, but they certainly happened on Alan Cave’s watch. They have certainly arisen during this period of Government. Our job is to look at allegations, whenever they relate to. I will look at one that goes back to 2001. If it is found that A4e has committed a fraud in any of these many cases that you are looking at, will you reconsider, at that point, whether it is a fit company to undertake the new Work Programme, whatever the controls you have on that programme?

Robert Devereux: These companies have got contracts with us. We have made it perfectly clear that, like you, we will look back as far as we need to examine some of these frauds. The question that we will be asked, as a matter of law dealing with contracts, is whether the action we take is reasonable. If I found that, in 2008, there was a case, which, by the way, we may have recovered the money for, it may well not be reasonable under law to take a contract off them for that. If, on the other hand, we find out that, despite everything we have been saying, there are fantastically systemic problems, which persist to this day, then, as a couple of Ministers have said, yes, we would take the contract off them.

Q158 Austin Mitchell: Are there alternatives available? Is there competition for these contracts?

Robert Devereux: This is an important point: there was competition for the market-competition to get on to those contracts. Remember, at the moment, every region has got at least two, and sometimes three providers. They are currently getting random allocations of customers-say 50% each. In three years’ time, if it turns out that provider A is doing better than provider B, they will get more work than provider B. Even in situ-even with this contract-it is a wasting asset if they do not perform.

Q159 Chair: I not worried about your not finding alternatives, because I agree that there is competition. I am interested: what is your definition of systemic? What is it that has to be shown for you to say, "No more"?

Robert Devereux: I will have to think about that, and drop you a note. To me, systemic means the very way in which the company is run.

Chair: That is what we have evidence on.

Robert Devereux: No, what you have evidence of is the past, so be very clear-

Q160 Fiona Mactaggart: We have evidence of company cultures.

Robert Devereux: Yes, but companies, like individuals, may change-

Q161 Mr Bacon: We have just established that they have not changed. Indeed, where the beneficial ownership had changed, you would have had to have been notified.

Robert Devereux: You are making a very simple elision between some past facts and contemporary culture, leaving out the fact that consequent upon those facts we have changed some of the processes. You cannot have some of the culture-

Chair: I know we cannot talk about it because it was in private, but the auditor who sat before us was there-they were both there-between 2011 and 2012. This was not massively in the past.

Matthew Hancock: That is not actually accurate.

Chair: They were.

Q162 Matthew Hancock: No, one left in May 2011, so was not there in 2012. But more importantly, as we discussed, this needs to be investigated. Do you agree? It should be investigated properly, and then you can come back to us.

Robert Devereux: I know what you will tell me, but that auditor did not work for an employment programme.

Q163 Chair: But he was within the organisation. This is where I have to say to you that while I understand, I simply ask you to work across government, because our interest here is that it does not matter whose contract it was.

Robert Devereux: Our consistent position has been that if you have any information let us know and we will put-

Q164 Chair: We have given you information, and I hope that you will come back to me and talk about the ones where you felt that you did not have enough information.

Robert Devereux: I will bring an investigator round.

Q165 Chair: Okay. Finally, a public sector contract to build houses, hospitals or schools is pretty straightforward. I am thinking of payment by results-the school is there-and you can keep a little bit of money back to make sure that if anything has gone wrong, you can get it back. Would you accept that a contract that provides human services is qualitatively different? It will work only if you build trust. I think that that is the key thing-there has to be trust in the system. It is partly about the outcome but also about-this is the Fiona point, really-how well the customers are treated. That has to be part of the mix in these sorts of contracts. So it is about the how as well as the how much-I hope that you accept that. If I put it to you that I think that trust has broken down here, in particular in relation to A4e, how can you rebuild it?

Robert Devereux: First, I agree that services to individuals are important. Any company that thinks it will be successful as a company if it is not having trusting relationships with its clients is probably making a mistake. The idea that somehow or other I can have a company that continues to be very successful in getting people into work but that actually has no constructive relationship with them I think is a bit difficult. Imagine for a minute, just hypothetically, that it was successful in getting long-term people into work without necessarily the friendliest relationships, I would still be quite pleased with that outcome, because I do not really want them on benefits and I would rather have them in productive employment. I agree that it would be nice if the relationships were positive, and I suspect that positive relationships add to their ability to get people into work, but that is sort of where I would pause. I do not actually think, from what I have heard, that lots and lots of A4e customers have lost trust. As I said to you earlier, 11 cases out of 750,000 referrals have made it to the second tier, and I can assure you that second-tier complaints are a really easy thing to do, because I read them all the time on the benefits side. There is a sense in this Committee, I fear, that we have decided beforehand that this thing is systematically wrong. You are nodding-

Matthew Hancock: Not everybody.

Q166 Chair: Within A4e.

Robert Devereux: No, across the piece. There is that sense here. All I am asking you to do is to be a little bit clinical about when the evidence you have seen is from, what changes we have made and what changes the companies may have made. By all means keep testing us on it, because we are not perfect, but I would like to end by concluding that, in one sense, how nice as an accounting officer to find that the NAO thinks that I have done the right thing. They do not always say that to me, as you know.

Q167 Chair: I don’t ignore the evidence that we have. This is probably a conversation for elsewhere, but I have to say to you, Mr Devereux-

Robert Devereux: Those are allegations that you have, not evidence you have.

Q168 Chair: The allegations we have-but the way in which the DWP tends to deal with these allegations, as compared with other Departments which are also making contracts with private providers, whether for educational or for health services, is qualitatively different. There is a greater openness to looking at the systems. That is my experience-

Alan Cave: What is the evidence?

Robert Devereux: What is the evidence?

Q169 Chair: That is my experience. You are never actually prepared to say, "Hang on a minute, this might mean that there is a bit of a system here that we ought to investigate and see whether or not it is viable."

Robert Devereux: If we had not thought that, we would not have invented the programme that is in front of you today. We could have sat on the New Deal; we could have sat on 600 providers, 1,300 contracts-

Chair: All political parties were going towards what has become known as the Work Programme . As I keep saying, it was evolutionary-

Fiona Mactaggart: That is evident from the report that I quoted.

Chair: Yes, it was no great revolution; it was evolution.

Q170 Matthew Hancock: Do you accept that it is not evident how it reduces fraud by requiring outcomes rather than processes? It is entirely outcomes-based.

Robert Devereux: The level of automatic checking was not the same as we have now. I am able to make these statements, with which I had some other things to say, because we have invested in the system that we have just turned on and which you keep asking about-it has just been turned on; it did not pre-exist.

Q171 Chair: I am not sure that it is value for money.

Robert Devereux: The value-for-money question goes back to our earlier hearing, because that is the point that Matthew made.

Prepared 27th September 2012