Public Accounts - Minutes of EvidenceHC 136

Back to Report

Oral Evidence

Taken before the Committee of Public Accounts

on Monday 11 March 2013

Members present:

Margaret Hodge (Chair)

Mr Richard Bacon

Guto Bebb

Jackie Doyle-Price

Meg Hillier

Fiona Mactaggart

Austin Mitchell

Ian Swales

Justin Tomlinson


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


Responding to change in jobcentres (HC 955)

Examination of Witness

Witness: Katie Shaw, Head of Welfare Policy, Citizens Advice, gave evidence.

Q1 Chair: Welcome, Katie. Thank you very much for agreeing to give evidence at the start. We had a colleague of yours here last week. This is rather a short session; it is simply for you to give the perspective of people who come to the CAB and their perceptions of how the Jobcentre Plus reforms are impacting on clients. Would you like to start by saying what key issues we should probe further when we get the accounting officer and others in?

Katie Shaw: Thank you for asking me to give evidence today. I will start by looking at what the NAO Report found and how our evidence and experience at the CAB relate to that. One of the main things was that, during the economic downturn, when the demand on services greatly increased, they focused on paying benefit and reducing some of the other initiatives, such as the work-focused interviews, the work search support and sanctions, so they managed to keep paying benefits on time. Our evidence would concur very much with that; we had a massive increase in jobseeker’s allowance inquiries during that period, but we did not get a huge volume of inquiries about delays in paying benefit in a similar way to a few years earlier, when they changed the way that benefit services were delivered. For us, how they choose to deliver services is one of the really key things. That is one of the things it is good to focus on. Going forward, our services are already beginning to change in terms of the way you claim benefit as we move forwards towards universal credit.

The other thing is that the focus on the majority, and on it working for the majority, often means that we forget or do not pay such close attention to how the minority are managing those changes. They are the kind of people we see. Those people are often the most vulnerable, are less able to help themselves and struggle with those changes, even when, as a whole, it is going okay.

Q2 Chair: I was going to ask you about that. Let’s take the parking issue first. There is a suggestion in the Report and in other evidence from research that Jobcentre Plus have done on themselves that, because of the pressure on resources, they are tending not to give the wider range of back-to-work services, particularly to ESA claimants who will be ex-IB claimants and IB claimants who have gone on to JSA-those sort of people. Is there any evidence from your work that that issue is emerging?

Katie Shaw: In our evidence, we see quite a lot of people who are sanctioned.

Q3 Chair: More than you did?

Katie Shaw: More than we did. Generally, people come to us when something goes wrong, so if they are really eager for support and they do not get it, they will come to us. Certainly, that is something we saw during the economic downturn: the decision to say to people, "Go home and claim on your own phone, rather than use Jobcentre Plus services" may have been an efficient use of services, but how it was done often left people a bit dissatisfied. They hadn’t been to Jobcentre Plus for many years, they’d made all that effort to go, and then when they got there they were told, "Go home, you are better off at home doing it yourself," without proper, detailed checking of whether someone could go home and do it themselves-that they had access online or had a land-line rather than a mobile. It is that discretion-that checking whether what you are saying is appropriate for everybody, or is just a general message-that is going out too generally.

People coming off ESA and people claiming JSA with extra support needs is probably one of the groups of biggest concern for us. We see a large number of people who are sanctioned. We have seen a 45% increase in sanctions over this past quarter, compared with the same quarter the previous year.

Q4 Chair: A 45% increase? That doesn’t come out in the Report, does it?

Katie Shaw: The Report says that they went down during 2008-09, then jumped back up again, and then have gone back down to the original common average.

Obviously, we are more likely to see people who have been hit with sanctions and whose sanctions have possibly been given unfairly. If people legitimately were not really trying hard enough and were pulled up by a JSA sanction, they would not necessarily come to us to seek help if they were able to help themselves.

Q5 Chair: But you have seen a 45% increase in the use of sanctions. Is that across the board?

Katie Shaw: Not the use of sanctions: people seeking advice about sanctions inquiries. That can be different. It has been a slow rise over the whole period. Obviously, JSA inquiries have increased overall and we must remember that context. From 2008-09 to the last full financial year, our inquiries about JSA increased by over 100%; that is a reflection of the economic position we are in, rather than necessarily of things going well or badly within Jobcentre Plus. Even when JSA inquiries overall slowed down-last year we saw a 2% fall in inquiries about JSA overall-we still saw an increase in sanctions inquiries. That has been fairly steady. A 45% increase when we have already seen an increase in previous quarters is quite significant. In the three or four months before last Christmas, sanctions got tougher. A full sanction was introduced for ESA claimants for the first time. Leading up to Christmas, it was tougher sanctions for JSA claimants, who could end up with a three-year sanction.

We have heard many times the Secretary of State say that the sanctions are not expected to be used except in a small handful of cases, but they are given precisely for the things that we see people come into CAB with every week and every day. They are where the claimant has failed to follow a direction, which means they failed to apply for a specific job or take a specific action. Our evidence shows that people who have learning disabilities or mild mental health problems often come in and say that they didn’t understand what was required of them, or were unable to do it because of literacy issues, or the claim was required to be made online and they were not able to do that.

That kind of understanding of why someone hasn’t complied is what we don’t see in the cases that people seek advice about. We are not yet confident that there has been that conversation that fully explores why they haven’t been able to do that. That is why we are concerned, going forward, that those safeguards with tougher sanctions don’t seem to be in place and are very much up to the discretion of the adviser.

Q6 Guto Bebb: Just on an anecdotal basis, I work very closely with my citizens advice bureau in north Wales, and there was a period when people were coming through regularly and I was supporting them in terms of CAB and sanctions. That has fallen off quite dramatically recently. What you are saying contrasts with what I am witnessing. Are there any figures for sanctioning over the past six months? Are you seeing less or more?

Katie Shaw: In the first two quarters of last year, it had fallen slightly compared with the same quarters of the previous year. But from September to December, we have seen a rise again-a 35% rise in the second quarter and 45% in the third quarter1.

Q7 Guto Bebb: Is there any regional variation?

Katie Shaw: There would be some. I would need to look at the regional figures and the Welsh figures.

I know that is something my office in Wales is concerned about. Whether it has fallen off slightly-it is just an ongoing issue. In the evidence they submitted about the Work programme recently, sanctions seemed to be the biggest issue they were concerned about in that area too. But it is good to hear if it is dropping off in your area.

Q8 Ian Swales: I know that my council has cut down on its welfare and benefits advice service because of the pressure on its budgets. Do you know whether that has happened elsewhere and whether that is affecting the figures you are talking about?

Katie Shaw: It is, yes-both welfare and benefits advice from the core funding for local authorities; and from April, we are very worried in particular about the huge drop in funding from the legal aid cuts. It will have an impact on our figures, and it is interesting, as we go forward from April, to look at the whole raft of welfare reform changes that will come in. We might not see rises in inquiries, because rises assume this capacity to deal with increases. That would be a big concern with how we report the impact on advice services and, more importantly, on claimants.

Q9 Ian Swales: In terms of the historical information, are you saying that you think you may have had more inquiries because councils have less resource, and then in future, you are concerned that your figures may not be accurate because only so many people can get through your door? Is that what you are trying to say?

Katie Shaw: They only reflect the people whom we can see, so they don’t reflect unmet demand. Some of our figures say that, overall, benefits have dropped slightly. Over the years from 2008-09, benefit inquiries increased every single year when overall inquiries have not. That has dropped down again now.

Q10 Chair: Has it dropped down as a proportion?

Katie Shaw: As an overall number. Benefits have risen as a proportion overall.

Q11 Chair: So it has risen as a proportion. Because you are able to do less work, they are taking a bigger slice of the cake.

Katie Shaw: Yes, that’s right. Debt was our biggest area of inquiry. Benefits followed shortly behind, and now it has swapped round.

Chair: Oh really?

Katie Shaw: Yes. Obviously, they are closely related. But you can’t keep reporting rises because, as I say, it does hit a plateau. We had extra funding for the first two to three years of the recession, which enabled us to help more people. Now that has plateaued, dealing with more benefit inquiries means dealing with less of something else, but they tend to be the more urgent ones.

Q12 Ian Swales: Figure 8 in the Report makes it clear that jobcentre resources are variable across the country, with a 30% difference between the staff in the most heavily loaded area compared with the lightest loaded. You may not have had time to check it, but do you see any correlation between that data and the pressure on your officers? In other words, in areas where the jobcentres are most hard-pressed, does that result in a bigger load for you?

Katie Shaw: I haven’t done the comparison, but I would not be surprised if that were the case. Certainly, we do get a lot of inquiries from people who have referred directly to CAB for help.

Q13 Ian Swales: From the jobcentres?

Katie Shaw: Yes. That is often for help with claiming. There is now a six-pilot trailblazer, I think it is called, for claiming JSA online, and we are seeing a lot of people who urgently come to CAB because they can’t claim online-

Q14 Ian Swales: This is something I find particularly interesting. A lot of people in the third sector, like yourselves, are concerned about lower funding. They come and see me, and I often say to them, "Well, talk to me about what you’re doing." In a case like that, you are actually providing a public service, no doubt for nothing. You are not getting money from the jobcentres to help people to fill in their forms, are you?

Katie Shaw: No.

Q15 Ian Swales: So this is something I think we want to explore with them. If they put a difficult system in place, they can’t then offload dealing with it to the third sector. That does not seem right, unless some money changes hands and you become their advisers or whatever.

Katie Shaw: We very much agree with that.

Q16 Austin Mitchell: Do they tend to come to you first or go to the jobcentre first, get baffled and come to you?

Katie Shaw: It is a complete mixture of both, to be honest. In the early stages of the recession, when people were needing help to claim for the first time, they would usually go to a Jobcentre Plus office; they would only come to us if they had difficulties. Often, there were people who were quite able to self-manage, but there are others who struggle with Jobcentre Plus services, particularly if it is about phone claiming. Some people will always need that extra support. But generally they have tried to deal with Jobcentre Plus first.

Q17 Austin Mitchell: What help do you provide them with? We gather that the proportion of people applying online for jobseeker’s has risen to 40% to 50%. Do you provide them with help in applying online?

Katie Shaw: Some bureaux no doubt will. You can still claim on the phone. Certainly some recent evidence from bureaux is that they have tried to help that person claim on the phone. If they are not able to manage the claim on an ongoing basis on their own, they will help them to do a phone claim instead. It is a mix, to be honest, but yes, they will help them to claim-

Q18 Austin Mitchell: So some centres will and some won’t?

Katie Shaw: Some will, yes. It will in part depend on their confidence in the ability of the claimant to manage the claim on an ongoing basis in that way-to interact with Jobcentre Plus through the internet.

Q19 Austin Mitchell: To pursue Ian’s point, you are providing something of a public service, in the sense that you are sending them into the jobcentre better armed as to what they need and how they should be treated. Is that correct?

Katie Shaw: Yes, I think that is possibly slightly more typical than helping them to claim. Most bureaux would strongly assert that the very basis of making a claim for jobseeker’s allowance or other benefits is something they should be able to do through help from Jobcentre Plus.

For example, in a recent case, the bureau had had a letter saying, "We’re going to make a push to encourage people to claim JSA online, and this will be the preferred route. Claims made online will be processed more quickly than those made over the phone, but you can still do it over the phone if that is what the claimant really needs."

The bureau knew this claimant could not manage an internet claim, so they sent them equipped with exactly what they needed to say and what they needed to ask when they made the call to insist that they should be able to make a phone claim, and the claimant still came back to them and said, "No, I was forced to make the claim online" or said they couldn’t manage it themselves.

You have to remember that somebody who struggles and is more vulnerable in expressing themselves or communicating is the least able to argue that they need extra support to do something. They came back to the bureau because they had not been able to do it, so then the bureau made a complaint to the jobcentre but also helped that claimant to make the claim over the phone.

The person would be referred, on the phone line, to their local library or to their CAB-somewhere they might get help-but it might be that the library is not necessarily equipped to do that. I was at a conference recently. The local authority person was saying, "But we’ve got one library with computers in the whole of our local authority area and there isn’t someone to help." That is the other thing you have to remember-that making a claim online may be possible for someone if they have help and assistance to do it.

Obviously, the Government are reinforcing the importance of assisted digital. But you can’t assume the two things are the same. Just because someone can do it with help doesn’t mean to say they can do it on their own and feel confident about getting it right. It is important information you are giving across.

Q20 Austin Mitchell: I get the impression, from talking to folk in Grimsby, that in the recession, because of the pressures, the services have got a bit more perfunctory, a bit more conveyor belt-like and a bit less satisfactory. Is that your impression, too, and have you needed to offset that to a greater degree?

Katie Shaw: In terms of the extra support that we provide?

Q21 Austin Mitchell: Their treatment is less personal and much more conveyor belt-like.

Katie Shaw: Within jobcentres or within the bureaux?

Austin Mitchell: In the jobcentres.

Katie Shaw: Well, yes. JobCentre Plus is encouraged to send people away to self-serve, to do it by the phone or to do it online, and there is less encouragement to use the warm phones even in JobCentre Plus or at the JobPoints. Hopefully, that will change-

Q22 Austin Mitchell: What I am asking is whether the jobcentre service has deteriorated because of the pressures.

Katie Shaw: I think they have focused on what they consider to be priority issues. The clients that we see would benefit from more support to move closer to the labour market. We see those who have had sanctions when they could have done with support. That would have moved them closer, but the sanctions have moved them further away, because their needs have not been picked up. If you punish someone when you have not picked up their needs, you will push them further away from the labour market rather than draw them closer. That has been our concern, but of course we do see people most in need.

Q23 Justin Tomlinson: On the point about digital engagement, I absolutely get everything you are saying, so I will try not to lead you too much in this question. Has the introduction of the money advice service helped at all?

Katie Shaw: I think it is difficult for me to answer that one. I am sure that it has provided some help to people, and it has referred people on to face-to-face support, which many people with money issues need. So, yes it does help, but it is only a part of the picture.

Q24 Justin Tomlinson: We also work very closely with our local citizen’s advice, and we are finding that those who are equipped to engage with the system as it stands are reasonably okay, though, sometimes, if it is their first time, they perhaps need a bit of support and help.

Those who find digital engagement challenging, who have everything in a carrier bag and who do not know what to do need that face-to-face support, which is where, I imagine, you have then had to take up the slack, where there has been an influx of extra people. I do not think that the money advice service offers that.

Katie Shaw: They do not offer that detailed support, no. I think that we cannot over-emphasise just what a big change moving to digital services will be for some people. The quality of service, hopefully-I am optimistic-will be good for those who can engage with it.

There will be much clearer and much more accessible information than people have had before from DWP in terms of letters and communications, but it is a long, long way for a lot of people before they get to a position where they are confidently using those digital services for private information and for detailed and challenging information.

Yes, we still see a lot of people with those carrier bags of paperwork, whether it be their debts or their benefits; they have everything together and they need help getting that sorted out face to face.

Q25 Chair: One of the issues is the relationship with Jobcentre Plus and the Work programme providers where you can see there is a bit of competition. The new system probably means that there is not an incentive for Jobcentre Plus to help the Work programme providers very much because they do not get any credit for it. From your point of view, looking at the case-load you get, has that been an issue? Have you picked it up at all?

Katie Shaw: What we have seen sometimes is an issue of administration and of clarity over who is responsible and who a claimant should report to. If they cannot turn up to an appointment, they might have phoned one and it should have been the other. There is that confusion of roles.

Again, we tend to see things when they go wrong in terms of sanctions. It might be that they have been asked to do something, but they thought that they had told someone else. It is just that confusion about who is in charge. They sign on at the JobCentre Plus office, but they meet with their Work programme providers. So a lack of clarity about who is in charge of their job search is one of the issues that we see.

Q26 Chair: The final thing is that they now have much more flexibility as to what to offer at the local jobcentre. Have you picked up any view as to how that is played out to clients?

Katie Shaw: I do not think we have, really. It is something that I have looked at, thinking that it is exactly what we have been asking for. It is hugely positive. It would probably mean an absence of inquiries coming into us about those issues that we would have seen otherwise. I do not think that it would be something that we would be very easily able to pick up except by doing direct research with our bureaux and clients about it. What I read about it is optimistic and it has the kind of approach that we have been looking at for them to be able to take.

Q27 Fiona Mactaggart: You said at the beginning that your impression was that the online application process had most heavily burdened relatively vulnerable applicants. Do you have any impression about who is most burdened by sanctions, and whether that same pattern is reflected in the sanctions regime?

Katie Shaw: From our evidence, the people we see who are sanctioned appear disproportionately from vulnerable groups-people who have come off incapacity benefit or who have sicknesses or disabilities, mild mental health problems, or learning disabilities, which in many cases would mean they never qualify for ESA and are capable of working, but would need significant support to do so.

We also see people with language difficulties; they have not fully understood what is required of them. We have seen cases where people have applied for several jobs, but not for the exact job that the Jobcentre Plus adviser had suggested they apply for. However, it was something very similar-the same sort of job with a different company.

We see people who have no idea why they have been sanctioned, or why their benefit has been stopped, until they come to us and seek advice and we are able to find out and get to the bottom of things. That is quite distressing, because how could a sanctions regime work if the people who are being sanctioned do not know about it in advance-so it is not a warning-and cannot avoid it? So all it can do is push people into hardship. There are some people we see who are really trying hard to find work, but because of their low levels of skills, or because of their disabilities or sicknesses, the job market will be narrower for them, and that is a real concern.

Q28 Chair: Is there anything else you want to add, which you think you have not been able to cover in this short exchange?

Katie Shaw: Really, it is about going forward, and looking forward to universal credit and to assisting people. Again, more people are claiming online, and there are people who will have to manage in different ways, and to claim in different ways to be paid. We would say that the main lesson is that things might work for the majority, but you really have to look at what is happening to the minority.

Now, 40% of JSA claims are made online. I really wish we could pick up how many people have tried to claim JSA and gave up. Figures do not really show that and it is quite hard to pick up, just as the NAO Report picked up the fact that off-flows are measured but not where those people go. That is crucially important, particularly to understand people who are more vulnerable and might give up and get lost. Those things are important.

Going forward, regarding the support that people will need under the new system, I would say that many people will need those alternative payment methods, which will be more frequent than monthly payments, on an ongoing basis and closely watched and monitored, and we should not assume that with a little bit of support from an online budgeting tool, or even from a face-to-face meeting, they will be able to manage a single monthly payment to a household.

Our concern would be that, yes, it is good to put that support in, but do it early, do it thoroughly and assess how it is working before you take those sort of supports away from people in terms of that weekly payment. That weekly payment enables people to budget, and taking it away, or saying that you can have a weekly payment but after a short time you will go back to a monthly payment, is really worrying, given that our experience in the past has been about how often the minority suffer when changes happen in the interests of the majority and of making efficiency savings.

Chair: Thank you very much indeed. That was really helpful and very clear.

Examination of Witnesses

Witnesses: Robert Devereux, Permanent Secretary, Department for Work and Pensions, and Neil Couling, Work Services Director, Department for Work and Pensions, gave evidence.

Q29 Chair: Welcome. This Report is not a bad read for you.

Robert Devereux: You must be happy. It says "value for money" at the front, so I am in favour of it.

Q30 Chair: It makes a change in our exchanges. I hope you see the questions in that context. I want to start with the issue of your use of off-flows, rather than job outcomes, as a way of measuring the impact of the work in Jobcentre Plus. Why did you do that? What has been the impact of that change to off-flows from job outcomes?

Robert Devereux: As the Report says, off-flows are a relatively intuitive measure of the numbers of people leaving JSA in any one period. That was a conscious choice from the previous world, where we used to have a weighted average, so that certain people would get more points than others. We used to chase points. Mr Couling actually knows more about this than me because he was doing it at the time. We are not now trying to calculate some equivalence between somebody who has been here for two years and somebody who walked on yesterday. We are simply interested in whether people are leaving and, because of the strong relationship between how long you have been on benefits and your likelihood of leaving-as you can see in figure 11-I do not think that it is obvious to me that the additional complication and apparent sophistication actually merited the calculations. This is a straightforward way of measuring something that we can all see.

Q31 Chair: But why did you do it?

Robert Devereux: Because it is more straightforward to understand. In order to do the previous one, you basically had to have the view-

Q32 Chair: Has it changed behaviour in Jobcentre Plus?

Robert Devereux: I think it has sharpened up what it is we are asking, which is actually that we would like anyone on benefits to be somewhere else-preferably in employment. That is where we have gone with it.

Q33 Fiona Mactaggart: Are people who are sanctioned counted as off-flow?

Robert Devereux: If they are still on benefits, they are still on benefits.

Q34 Fiona Mactaggart: In my experience, if they are sanctioned, they do not get their benefits. That is what those who ask me for help say.

Neil Couling: They are sanctioned on the JSA element of their claim. For example, if you have JSA plus some other elements, you are just sanctioned on the single person’s entitlement of the JSA. It is possible to stay on JSA and be sanctioned.

Robert Devereux: It depends on what your household make-up is.

Max Tse: There are two types. There are sanctions, which are often fixed or variable term, and you stay on JSA, but just lose it for a bit. There are also disallowances, where you lose the entire benefit.

Q35 Mr Bacon: Does the latter count as off-flow?

Robert Devereux: Yes. If you are not receiving the benefit, you are not receiving the benefit. Correct.

Q36 Chair: One of the interesting things is that, according to the Report, you do not know where they flow off to. According to the Report, you have no idea where 40% of people have gone. They disappear.

Robert Devereux: The Report says that we have two ways of tracking them. We have ways of understanding what claimants are saying to us. Some people who get a job come back and tell my staff, "I’m really pleased with what you’ve done. I’ve got a job with Asda." Other people just disappear. There is a bit of a theme running through the Report, which is about the only value of jobcentres being to help people into work. The whole point of the JSA regime is to place a clear obligation on the claimant to be looking for work and to be trying to get off benefits themselves. Actually, I am genuinely and perfectly happy for them to leave benefit, because some of those people who are leaving-though it may be a small proportion-should not have been there in the first place.

Q37 Chair: I accept all that. However, accepting your assumption that the most important thing is to cut the benefit bill, surely it would be helpful to you to know where the 40% who disappear go to.

Robert Devereux: I think all we are arguing about is whether or not I have an elaborate administrative system as opposed to periodic, survey-based data.

Q38 Chair: Do you accept what I said? Do you accept that it would be more helpful to you, if over time you want to control the benefit bill, to know where they go to?

Robert Devereux: I am not sure I need to know where every claimant goes to be able to run the system. I have a set of figures, as in figure 13, that says, on a sample basis from 2004, 2008, 2009 and 2011, when we actually went and tried to track down individuals, we are getting a very consistent view that two thirds of the people who leave us are going into work and a third going elsewhere, which is either on to another benefit or something else. But we know that two thirds are going to work. In the old days, I used to have some quite expensive tracking teams, so if I referred someone to a vacancy, an administrator would ring the employer up and ask: "Did Mr Couling go for that job? Did he get that job? Has he started?" Unsurprisingly, employers found this an incredible overhead: the staff cost is quite high and, at the end of the day, it does not tell you anything better than the periodic survey.

Q39 Ian Swales: But you already have some extra data, because you know who you have sanctioned and you know where you have pursued benefit fraud cases. You already know that it is not simply "work" or "unknown" from your own work, don’t you?

Robert Devereux: I thought that what the Chair was asking me was whether I feel disadvantaged by not having management information of where everyone was going. I do not feel so disadvantaged, because I feel that I have enough information from basic, good-quality research to know the answers to those questions.

May I say one more thing while we are at it? This will of course change in a world of universal credit. At the moment, if someone goes into low-paid work and is with the Revenue, all I know is that they have gone into the Revenue. I may do some data-matching after the fact. In a world in which they go into low-paid work and they are still on universal credit, they will still be one of my claimants. I will have a whole bunch of people on the books who have either no earnings-typically, out of work-or some earnings or a bit more earnings. So the information flow within the system will be transformed by universal credit, and that I regard as being a much more profitable line of interest.

Q40 Meg Hillier: You just mentioned, Mr Devereux, one of the things that I was puzzled about. If you have got people going into work, because they will most likely be PAYE, surely you can cross-match on their national insurance number whether they are working? Surely that is a fairly automatic match?

Robert Devereux: I could do, but that is not a standard part of the MI or management information basis for in the jobcentres. I have survey-based information and some conception of how the economy in the labour market works, but I do not regard it as changing the price of fish in the Evesham jobcentre to know precisely where everyone went.

Neil Couling: That is what we used to try to do. Between about 2005 and 2009, that is how we tried to measure Jobcentre Plus performance. It was fraught with two difficulties. First, there were two different sets of data from HMRC-one relating to the P45 data and one to its general records of national insurance and tax-and they did not match up. There were great time lags as well, so we were trying to drive a car by looking in the rear-view mirror each time. We moved to off-flow therefore because it is a much more visible and immediate thing for managers and indeed staff in the offices to work with.

Q41 Meg Hillier: With real-time information coming in HMRC, will this make it easier to do what we were doing? Is that something that is possible to do? Will the system allow that?

Robert Devereux: For real-time information, in a world of universal credit, I will know monthly for all those people who move into PAYE-not into self-employment-exactly what their earnings are. I have to know that in order to be able to compute the universal credit, so the data source becomes much richer, much quicker.

Q42 Meg Hillier: So you will have much more knowledge about the 40% who at the moment fail to sign on or are not known? You will know much more about a percentage of them.

Robert Devereux: I suspect that what I am actually going to know is a lot more about the 60%. For the two thirds going into work, because they are PAYE people, I will be able to know that positively, whereas my current administrative data typically underestimate that by 10% to 20%.

Q43 Meg Hillier: With the ones you do not know about, how many is it? Do you track people when they move areas? If someone left Hackney and went to Redcar, would you know that that was why they were not signing on in Hackney? Because, presumably, they would still be in the DWP national system, they should show up.

Robert Devereux: They would. It would depend a little bit on whether or not they literally stopped claiming in Hackney and went somewhere else, or whether they simply got their claim moved. I believe that they can do either of those.

Q44 Meg Hillier: If they stopped claiming in Hackney and turned up in Redcar the next day to claim there-

Robert Devereux: It would not be apparent at the moment, no.

Q45 Meg Hillier: So they would count as off-flow in Hackney?

Robert Devereux: And as on-flow somewhere else, yes.

Q46 Meg Hillier: So there is quite a lot of unknown potential. Just picking up on Mr Couling, you were saying that you used to ring people up to find out about their jobs. Does any of that go on now? If I went to an interview, would you have a percentage of people who your advisers ring up to see?

Neil Couling: No, because the feedback from employers was that they did not want to work with us and put their vacancies with us, because of the aggressive follow-up, as they saw it.

Q47 Meg Hillier: Aggressive?

Neil Couling: We would be chasing them, saying, "We submitted 16 people to you. How many did you take on?" They may well have already got rid of their interview records and so on.

Q48 Meg Hillier: I can see, from the point of view of employers, that you have to work in a practical real world and there might be an issue there. There is also surely an issue for staff training and support, especially if you have someone long-term unemployed. Now they would be with the Work programme. However, it is a way of monitoring what the challenge is for Meg Hillier from Hackney in getting a job, and why she is repeatedly not getting one, or whether, in this case, the Work programme is not delivering for somebody. If you give feedback-that they turned up and looked scruffy or were late or something like that-that would be useful.

Neil Couling: We do that slightly more subtly now. I have about 2,000 employer advisers who try to maintain relationships with employers. Rather than chase down individuals by national insurance number, that relationship with the employer will mean that the employer will say, "I’m sorry, but I don’t think you are submitting the right people to my vacancies at the moment," or, "The people you sent me last week are really good. I would like some more of the same sort." We got rid of follow-up like this and we are trying to build a more positive relationship with employers, because they are vital to our work in terms of placing people.

Q49 Meg Hillier: To pursue that a little bit further, I have an organisation in my constituency-believe it or not, in the middle of EC2, an internet company-that has a huge warehousing operation on three floors and has employed 80 extra people in the past year. To my horror, they get them through an agency in west London. I am trying to get that changed, so they get them somewhere in east London. Presumably, that is something that the jobcentre would be willing to place people into. If that is the case, as I imagine it is, how come that has not happened? Is there any way that employers like that can be encouraged to come to the jobcentre because local people are more likely to be applying for the jobs?

Neil Couling: It may not have happened in the past because employers might not have liked the previous relationship they had with us, because of the follow-up issue I was talking about a few moments ago. Yes, we would like that employer’s business. Afterwards, maybe we can speak about how we can facilitate that. I am doing quite a lot of work at the moment with Business in the Community to try to encourage employers to take unemployed people for their vacancies.

Quite a lot of firms have employment practices that they do not realise militate against unemployed people. For example, they have knock-out factors at the screening stage that say if someone has any gaps in employment they won’t take them. That is why work experience is really important because it allows people to fill gaps in CVs and show that they can work. So, perhaps afterwards we can talk about that.

Q50 Chair: I want to stick to the job outflows. I have got Fiona, Austin and Ian. Before I go to Fiona, I am trying to look at whether you have all the intended consequences known when you measure job outflows. For example, if the emphasis is on getting people off benefits, do you know how many go through a revolving door? That is, come back, let’s say, within three months? Do you monitor that?

Robert Devereux: We do. If you turn to paragraph 2.6, you will see that the NAO got to the same worry, that if we set targets at 13, 26, 39 and 52 weeks, would that be some sort of incentive for my staff to try particularly hard at 12 and a half weeks and 25 and a half weeks? Were that the case, the curve that you see in figure 11 for every one of the past four years would have spikes in it, because the offload, just at the magic number, would actually rise. As you can see, that curve is as flat as a pancake. So it does not look to the NAO or to me that there is any unintended consequence by fixing on particular dates. It looks as if-and this is the truth as you know from meeting any of my staff-they are passionate about doing this, so they get on and do things. The targets are there to ensure that we know what is going on, but there is no evidence in that picture that those four particular points are driving performance in a perverse way.

Q51 Chair: I accept that. I think I was asking a slightly different question. The flow offloads remain static, but out of those who flow off, how many come back on again? The case may be encouraged off, or it may be somebody with a mental health problem. It is usually with those sorts of issues that they are encouraged off, and then they actually do not cope and they reappear-I probably should have not taken an extreme disability case, but I am interested in the revolving door.

Robert Devereux: You are absolutely right. If we were pressing our staff and saying, "The only thing that really matters is off-flow", all kinds of clever tricks would turn up to have them off-flow, and they would just come back again. I am afraid that I have not brought with me the rate at which people do reclaim-which, of course, is dominated by people trying short paid work and that work then not working out, so we should not take the rate of reclaim as being evidence of that, but you are right to spot it. We are trying very hard, in working through the leadership of every jobcentre, to make sure that they realise that that is not what we mean by successful off-flow. You can tell, as you go around, whether people are just cooking the books, and advisers know whether they are being invited to cook the books. They do not want to do that; they joined this organisation in order to get good outcomes for claimants, not to cook the books.

Q52 Chair: Do you have the stats? Could you let us know?

Robert Devereux: I could let you have it. I am just not sure whether we have brought it-I haven’t brought it with me, certainly.

Neil Couling: I might have it somewhere. The labour market is incredibly dynamic. There are anything from 250,000 to 300,000 people each month flowing on to JSA and off it, so it is a very dynamic beast.

Q53 Chair: I do not know what dynamic means in that context.

Neil Couling: Well, a lot of people are moving around, which is actually why I think you asked Ms Shaw earlier whether there was competition between the Work programme and jobcentres. In fact, there isn’t, because if somebody who is with the Work programme gets a job, then ceases to be in employment, and then comes back and claims Jobseeker’s allowance, they remain part of the Work programme but count inside the jobcentre’s off-flow targets. So the incentives are to work with the Work programme and jobcentres-working together there. It is not a case of static-

Chair: We’ll come back to that later, because I am just trying to get the concept. It would be very good if you could let us have a note on how many you think there are. What proportion of your people coming off benefits re-emerge within the three-month period and then within the six-month period? Because I think they are a different bunch of people.

Q54 Mr Bacon: I am astonished that you do not know that off the tip of your tongue. I think the National Audit Office does and it is quite surprising that you do not. If it really is, to use your words, as "incredibly dynamic" a beast as you suggest, with a quarter of a million people coming on and coming off, one of the first, most obvious things you would wonder is how much of that is the same people. Are you really saying that you cannot tell us?

Neil Couling: What I am saying is just how material is that to the work of jobcentres?

Q55 Mr Bacon: Quite. I don’t know, and nor apparently do you. Is it 40%? Is it 3%?

Neil Couling: I am not sure it is because-

Q56 Mr Bacon: You are sure it is what? Material?

Neil Couling: No, I am not sure it is material, because it is about the dynamism of the UK labour market.

Q57 Mr Bacon: It might be, but we do not know to what extent-because there are so many people going into new jobs, or because there are so many people going into jobs that do not work and are coming back. The number we are looking for is this: of the quarter of a million people who you are describing each month, how many are the same people coming back? Correct me if I am wrong, but that appears to be what you cannot tell us off the top of your head, whereas to me, that is a really obvious question that you would want to know the answer to.

Neil Couling: Some people will be in jobs for a very short period, and some will be in jobs for longer periods, and so forth. We can give you that kind of breakdown, but the Department for Work and Pensions does not run the UK labour market.

Q58 Mr Bacon: I did not say it did. I am talking about the ones who you are dealing with because they are jobseeker’s allowance claimants-and they are not, and then others who are because they come back. Mr Tse, what is your estimate of this? Am I right that it is 40%?

Max Tse: Of the people who came into jobseeker’s allowance in 2011-12, around 60% had claimed in the past two years.

Chair: Around 60% have claimed-

Max Tse: Around 60% who had come in have claimed in the past two years.

Mr Bacon: Had previously claimed.

Max Tse: Had previously claimed-exactly. It is an older statistic, but of the people who then leave jobseeker’s allowance, around 40% subsequently come back within six months.

Q59 Mr Bacon: So I was right when I said 40%, off the top of my head. How is that I have got it off the top of my head, and you have not got it off the top of yours? You run this thing.

Robert Devereux: Yes, but I do not come with every possible statistic for every possible-

Q60 Mr Bacon: I would not expect you to, Mr Devereux.

Robert Devereux: You are effectively, because you have chosen one that I do not have, so my apologies.

Q61 Mr Bacon: I chose a really obvious one. It was being discussed because it was so obvious. Anyway, the answer is 40% apparently, but we would love a more detailed note.

Amyas Morse: We bring these up, not because we are trying to pick up things that you might or might not know, but it does sound like quite a meaningful number. Don’t you think it is?

Neil Couling: What kind of number would you want?

Amyas Morse: I am more interested, not in that you should know what it is, but in that you should agree that it would be a good idea to have a handle on it.

Neil Couling: I do, and we do have the number, but I do not think it drives the organisation.

Amyas Morse: I did not say that it did.

Neil Couling: We have to respond to how the UK labour market is. Our job is to try to help-

Chair: Let me interrupt you. The reason we picked this up is that what drives the organisation is reducing the number of people out of work who are dependent on benefit, because you want to cut the benefit bill. If you know that a certain bunch of people keep coming back through a revolving door, that ought to drive your organisation because it might mean that you take a different approach to those particular people to stop them coming back. It is not just a labour market issue-it may also be an issue about the sort of people they are-so it might be about training or supporting them in work. I have no idea what the answer is, but I am interested because I want to cut the benefit bill. I do not want those people to keep coming back to my service and therefore requiring benefits again.

Q62 Mr Bacon: To be honest, I was not just trying to pick holes and be clever by finding a number that you do not know; it seems to me so central a thing to want to know that it really amazes me that it is not at the top of your mind, as the Chair says. You are called the Department for Work. If people are leaving the benefits that you provide because they have work and then the same people come back, or if only 2% or 90% of the same people are coming back, surely that is extraordinarily interesting information on how what you are doing in the broader sweep of things is working and whether it is helping towards sustainable outcomes.

Robert Devereux: Up to a point, because in that set of observations there was no reference to what the nature of the British labour market is like. If, hypothetically, there are a lot of short-term roles, a lot of people will cycle back through Neil’s organisation through no fault of his. You are right to say that, if it is actually my staff who, through some jiggery-pokery, are taking people off and having them come back again, there would be a problem. As I have just established by showing the Chair the graph, there does not seem to be such jiggery-pokery going on.

Q63 Mr Bacon: By the way, I did not say that, if they are coming back, it is necessarily Mr Couling’s fault. I am sure he is a fine chap, but I was just saying that this is such interesting information that one would expect that you would have it.

Robert Devereux: By the same token, if we are going to trade information that we are surprised the other person does not have, nothing in this Report even starts to answer the question of whether the labour market regime is effective. Yet, as all the evidence suggests-figure 19 purports to show how poor the Department’s initial forecasts were on the recession-the reason why our forecasters, who are very good and who have a very good reputation in Government, predicted the recession having such a high impact is that all past recessions had much stronger unemployment rates than we have now. There is something about the labour market regime introduced by JSA, which jobcentres are operating, that has had the effect of materially reducing the number of people who are unemployed and the cost of that bill.

Q64 Fiona Mactaggart: How do you know it is the labour market regime and not a response of either workers or employers?

Robert Devereux: It is a combination of both. Because nobody knows-the Institute for Fiscal Studies does not know, the Government do not know and the Office for Budget Responsibility does not know-the National Audit Office felt unable to give us any credit whatsoever, but that seems intuitively improbable if we have a regime that is much more active in asking people to look for work and is chasing that up with sanctions that are much more effective than they used to be.

By the way, our caseload now has many more people who were previously classed as inactive in the benefit system. We have single mothers with older children, ex-incapacity benefit claimants and various other people going into the regime. Neil’s teams are arguably working with a more difficult group of people than they previously worked with in terms of their labour market prospects, yet, as can be seen from the picture we have just looked at in figure 11, the off-flow rates are holding up. The absolute level of unemployment is much lower than it would have been in any previous recession. That seems to me to be a serious outcome-related measure about which I would have thought the Committee would be interested, and it is what I came prepared to talk about, rather than the rate.

Q65 Jackie Doyle-Price: I agree with most of what you have just said. Obviously, for people who have a difficult job search, temporary contracts will ultimately be a way for them to get a long-term job outcome. You would expect for there to be some degree of churn, but the figure is quite large, and you also referred to self-employment earlier. Have you made an assessment of how many of those 40% who come back within six months are people who are perhaps working in the black economy and claiming benefits until it gets a bit too close so they withdraw and come back? Is there any way of analysing that?

Robert Devereux: Again, I don’t have the first figure, and I certainly don’t have the second figure, so you will have to let me go away and see what I know about the make-up of the rapid reclaims. One of the reasons why I was unapologetic in saying that off-flow is a good thing regardless of destination is that some of them will be in precisely that position. The more that advisers make it difficult for people to work and claim by requiring them to look for work and attend things to support them, the better the answer will be.

I was in Harrogate jobcentre the other day-I recommend it if anybody is passing-which had a really good story about giving front-line staff lots more space to be innovative about the offers that they make to people. They have worked out how to get people down to the theatre, and they have worked out how to get people into retail-lots and lots of offers. The flip side of their having so many local offers that they believe in is that they have greater confidence in sanctioning people who do not participate. That office has better offers than most places, and it sanctions a higher proportion of the caseload than any other part locally. Guess what? Their claimant count has fallen like a stone. They also have one of the most engaged teams in the north-east.

Q66 Chair: Before we get completely carried away on figure 11, if you look at 2009-10-at the height of the previous recession-you were at 91% to 92% off-flow. In 2011-12, when the labour market appears to be going in a different direction in the monthly stats, you were at 86% or 87%-I can’t quite work it out. So the percentage has gone down, to be fair. I do not know what that means in total numbers. That is before we get too carried away.

Neil Couling: Unemployment is basically flat, so there were 1.4 million on the JSA count across 2010 to 2012.

Q67 Chair: But I am saying that 2009-10 is when we were at the peak of the previous-

Neil Couling: Of that 1.4 million, which is pretty constant, the make-up has changed. The 1.4 million stayed like that, but in the make-up there is an additional 150,000 people who were formally on incapacity benefit or are lone parents on JSA. So we have maintained pretty much the same off-flow rate with a much trickier caseload to place into work.

Q68 Chair: It is not pretty much the same. You are down 5%.

Robert Devereux: With a 10% harder caseload. It sounds like a profit to me.

Q69 Chair: All I would say is that it would be really interesting to have a better understanding of what is really happening. We are both making assertions, but if one side had a decent analysis of it, it might be grounded in fact rather than conjecture.

Robert Devereux: I am trying not to make an assertion. I was using the graph to demonstrate an answer to an earlier question, and you have asked another one about the regime before the last election, what happened with long-term unemployment and whether or not their claims were broken. That was also a different arrangement, so we would have to go back over that too. When people went on to training allowances after certain points, their claim was physically broken. Consequently, you could not be on benefit for as long as you otherwise could be. That will manifestly have the effect of-

Q70 Chair: It would be interesting to get a proper understanding of it. On the wider issue, I can accept that you do not want an administrator pursuing every claim, and you will get better when you get to universal credit-I understand that. But if you really want to test whether your interventions are working, you should work a little bit at understanding it.

Robert Devereux: Well, we have done crate loads of analysis on this. Let us give you something about those few years. You can see that the difference is not in the early months; it is in the longer durations. There are elements in there that I am fairly sure are policy-related, and not to do with the efficacy of the jobcentres, but let us write you a note to explain it.

Neil Couling: The best bits of data are out of the benefit counts-how many people are actually on benefits. What people tend to focus on is the fact that unemployment is broadly flat, but if you look from 2010, there are 230,000 fewer people on working-age benefits. What is going on is that the number of people on benefits is coming down, the number of people that jobcentres are working with is about the same and we are having about the same success rate with a much more difficult caseload.

I think that is a really strong success story, but we will do you a piece of analysis that will set that out for you.

Q71 Fiona Mactaggart: Mr Devereux, you started talking about the sanctions regime, and I want you to tell us what it is for.

Robert Devereux: What the sanctions regime is for?

Fiona Mactaggart: Yes.

Robert Devereux: From first principles, JSA is a conditional benefit. The claimant has got obligations to be looking for work and to be actively seeking work. The sanctions regime is there in the event that those things are not happening.

Q72 Fiona Mactaggart: One of the things that we heard from the CAB, and that I hear in my constituency a lot, is that the people who are sanctioned often do not understand why they have been sanctioned. Do you think that is sensible?

Robert Devereux: If it were true, it would not be sensible.

Q73 Fiona Mactaggart: You think the CAB might be lying?

Robert Devereux: I think it is quite conceivable that since we are dealing with an organisation with 1.5 million different people who claim, there are bound to be some individuals who will struggle with this.

One of the things that we are doing-again, in preparation for universal credit with the claimant commitment that goes to the heart of it-is being much more explicit right at the start about what these conditions are and what the consequences for you are. If you look at a jobseeker’s agreement at the moment, that is not made as explicit as it will be in the claimant commitment. We have been trying this already, and it is going very well in terms of people’s perceptions and comprehension. If there is any shortfall, the very standard fashion in which the claimant commitment will operate would seek to address that.

Q74 Fiona Mactaggart: There are obviously groups of people who will find it easier to understand the consequences of particular behaviours, because their English is good, because they read well, because they are confident about using the internet and things like that, all of which will mean that they are more likely to comply with behaviours that you expect from claimants.

Do you do any analysis about what kinds of people are most affected by sanctions-whether they are people who do not have English as a first language, or people who have mental health problems and so on? Do you have any figures in the Department about that?

Robert Devereux: I do not know. I would have to go and check for you.

Neil Couling: We do track sanctions. We are quite keen to avoid any misunderstandings that there are targets attached to these-

Q75 Fiona Mactaggart: Ministers have promised me that there are no targets. I have to say that people who work for you keep telling me that there are targets, but there we go.

Neil Couling: We do have some understanding of the needs of people, for example, who do not speak English. I was in Glasgow-Laurieston-recently, where an adviser and I were interviewing a Slovenian national who was here under the EU free movement rules, and who had very little English. We were carefully translating the jobseeker’s agreement and checking back with the translator that the claimant understood her responsibilities, what she needed to do and what evidence she needed to turn up with the next time she signed on.

My folk do some rather painstaking efforts here to try to make sure that we apply the sanctions regime in a fair way. I have always emphasised to them that it is designed to help people get jobs, because it reminds them of their responsibilities. The only way you are going to get a job is to apply for it, so if people are not applying for jobs and if they are not signing up to the conditionality regime, they are going to fail because they will not get jobs.

Q76 Fiona Mactaggart: So do people get a warning that they are about to be sanctioned?

Neil Couling: People do, yes.

Q77 Fiona Mactaggart: Always?

Neil Couling: Not in every circumstance, but my advisers will say to people, "Your job search in the past couple of weeks has really not been up to muster." I was in a jobcentre in Barnsley recently where the adviser was quite careful to say to the individual, "Look, you have not fulfilled the terms of your jobseeker’s agreement. I understand that you might have misunderstood that, so I am going to let that go this time, but in two weeks’ time I expect to see that you have made four applications."

Q78 Fiona Mactaggart: Mr Couling, I am absolutely certain that jobcentre advisers do that when the boss is there. I have no doubt at all about that. I think that they usually do it when the boss is not there, but I do not think that there is a process that guarantees that that happens-that an applicant knows that their behaviour puts them directly at risk of being sanctioned.

Robert Devereux: Let’s try this, then. I have got 30-odd thousand staff trying to do this, and 1.4 million people on the count. At the end of the day, this is all going to get down to whether I, sitting opposite a claimant, am communicating well. No amount of process definition in the world will get around that.

We must trust that well-motivated public servants who are trying to do the right thing by people and understand the rules are having those conversations. The trouble is that you have only to find one to demonstrate that this is not happening, but it is not-

Q79 Fiona Mactaggart: I am not looking for individual errors. I understand that any system has individual errors in it, and you seem to think, Mr Devereux, that I want to pick on you. I absolutely do not, but what I am concerned about is that you have completely devolved to local offices how they do these things, as I understand it, although-

Robert Devereux: No, that is a mistake. On the sanctions regime, because of the consequences that flow from it, we are very clear about how it should operate.

Q80 Fiona Mactaggart: In the parity of operation there is no requirement for a written previous-

Robert Devereux: There is. There is a jobseeker’s agreement that the claimant will sign and will be talked through by the adviser at the new claim interview.

Q81 Fiona Mactaggart: Absolutely, but that happened three months ago, and this month I am not applying for those three jobs you pointed me at.

Robert Devereux: The whole point of the fortnightly regime is to bring people in, to remind them of their obligations and to check that it has happened. If we say even so they may not know, so in addition to everything else we must have a written warning of a possible sanction, is that really the regime you want to run?

Q82 Fiona Mactaggart: I think it probably is.

Robert Devereux: Okay.

Q83 Fiona Mactaggart: If the point of a sanction is to change behaviour rather than to save money, yes I think it probably is.

Robert Devereux: But you were kind enough earlier to say that you thought that most of the time they were probably doing it properly, so you are asking about a process point to make it secure. I am not persuaded personally that the elaboration and cost involved in everything you have just described would be worth it for a problem that may occur in some cases, but I am not sure is systematic, given the staff I know.

Q84 Fiona Mactaggart: The problem is that the staff you know tell me that they are given targets for sanctions. Ministers tell me they are not given targets for sanctions, and have given me their absolute commitment that they are not. So something is happening locally which is not what nationally we have agreed should happen.

Robert Devereux: Just try this thought experiment. Imagine you are the manager in a particular office and you can see that many of your advisers are sanctioning at a particular rate and by and large it is 5%, 8% or something like that, and Fred in the corner is doing 2%.

Q85 Fiona Mactaggart: But if Fred is doing 2% and has a higher into-work ratio than all the others, then let Fred carry on with his 2%.

Robert Devereux: Okay, so what you are down to is managers making judgments, but one of the things they would look at-this seems to me perfectly reasonable-is do you think it is easy to do a sanction. It is much easier to just let it wash. People must put themselves into a difficult place on the part of society to do a sanction, so all the manager is trying to do when looking at the rates at which people are sanctioned is to try to think whether that sounds reasonable.

Q86 Chair: Mr Couling, can you tell us which offices do the most sanctions? Do you have that management information?

Neil Couling: Yes, we do.

Q87 Chair: Do you use it?

Neil Couling: Not to go round to other offices and say they should be at this level, or that.

Q88 Chair: What do you use it for?

Neil Couling: I do it mostly because people ask parliamentary questions about it, and if I don’t have it I am told it is terrible that I don’t have it. What I want to try to get across to people is that there is no right level.

Q89 Chair: It would be very interesting if there was a massive difference between the use of sanctions in Barking, Hackney and anywhere. It would be interesting.

Neil Couling: Sanctions are not an outcome. They will be a function of the nature-

Q90 Chair: If we get a feel from our local jobcentres that there is some pressure in some jobcentres, you ought to be able to pick that up as management information. I just wonder whether you do, and whether you act on it.

Neil Couling: I do, and I have done. There was one particular office where I felt there was more sanctioning going on than would be natural in a normal labour market. That was a subjective assessment, because there is no right level of sanctions, but they were sanctioning so many people that I thought they must have been wrongly sanctioning them. I knew that because the interesting thing to track back is the success of the sanction. The jobcentre-my folk-put the person in for a sanction and it goes to a decision maker.

What was happening in that office was a very high overturn rate, when our decision-makers were saying, "There isn’t the evidence to sanction this person-their excuses for not having complied with their jobseeker’s agreement are good cause." Different labour markets will have different levels of sanctions underpinning them-there is no right level.

Q91 Mr Bacon: If you have an "unusually" high level of sanctions-to use Mr Devereux’s example from Harrogate-but that is combined with an unusually high level of engagement with the local labour market, so that you may expect there to be higher sanctions because so much is being done to help people into work, and it is working, you would not necessarily see a very high level of sanctioning as a problem.

I have a couple of questions. First, where that is the case-to take your Harrogate example-should we assume that when it went to the decision-maker, there was not an unusually high rate of rejection or turnover, because they were found to be valid sanctions?

Robert Devereux: You should assume that, yes.

Q92 Mr Bacon: The thing that really interests me is how you capture and spread out the Harrogate experience, if I can call it that. There is quite a lot in the Report about the fact that you are trying pilots in different places and working to assess the freedoms and flexibilities that local jobcentres are being given, but there is quite a bit more that you could do.

Indeed, you have various things-see paragraph 2.29 of the Report-a "‘bright ideas’ portal" and even a "‘dragon’s den’ forum", although that was closed down "because it was found to be slowing implementation of flexibilities and discouraged sharing of ideas", which is perhaps surprising.

The end of that paragraph says that during the NAO’s visits, it "found that staff shared good practice through informal networks within their existing jobcentre districts rather than more distant offices." That was the way things tended to happen, and I suppose that is obvious when you think about it. If there is something really good going on, how do you then capture that systematically and ensure that, if it is working, it gets to more distant offices, rather than someone just talking to the person in the next office because they know them?

Robert Devereux: A couple of things. In this organisation, we are positively trying to get ownership right at the coal face about the way in which people act and operate. We do not want everybody in the front line to feel that their actions are entirely dependent on Mr Couling making some decisions back in Whitehall.

The whole march of history about how the regime has changed, which is set out very nicely in one of these charts, has gone from being, "It’s Friday, it’s 13 weeks-I’m going to see you for 40 minutes," through to "I’m going to look at Mr Bacon’s CV and think about it. I’m going to do something for you and then one of my colleagues will do something else." That march of history is actually quite a big step and change for my organisation, and the fact that they have now got as far as making sure that they compare notes locally means that we are at least getting somewhere with that.

You are absolutely right that we then need to ask whether there is any evidence that something is actually common across the piece. However, one of the things that I think we will keep coming back to is that this idea that there is some standard, perfect information model-that so long as I knew all the variables, I could go to Harrogate, hold it up and say, "Do this"-is not going to work. The Harrogate labour market is unlike any other. The claimants in Harrogate are unlike any others, as are the interactions with the advisers. You need to be alive enough to realise that there may be hints and prompts in what other people are doing, but not so daft as to then say, "I’m now going to do those as standard practice."

Q93 Chair: I do not agree with that, and neither does the Comptroller and Auditor General, so I will let him come back on you.

Amyas Morse: I am not going to come back particularly forcefully. The suggestion in this conversation is that one wants everything to be done on a standard lockstep basis, which as far as I know is not a suggestion being made by anyone. On the other hand, there is some value in having good management information as well as statistics.

I want to ask a factual question. Given the changes, which we acknowledge in the Report, it is probably useful to have some information that allows you to keep track of whether there are any concerning or undesirable effects. Let us take for example the sanctioning of what I will call vulnerable groups-do you have a way of telling from the centre whether there is any unexpected uptick in vulnerable groups? I hasten to say that I am not telling you that you should have a standard measure for them. There are a lot of things I am not trying to say; I am just asking a question.

As you move on to online and the various other measures that you are putting in place-quite appropriately, in my view-if you suddenly find that something is inadvertently starting to produce a result that you didn’t want, do you have a leading indicator or some way of monitoring that that lets you know? If so, how does that work?

Neil Couling: Essentially, there are a whole range of data sources that you will look at, as a manager. That is why we have avoided having targets. The old regime had lots of targets and tried to calibrate off those. We have stuck with off-flow and we have a range of management information underneath that that causes you to ask questions. They do not provide you with the answers, but they cause you to have a discussion with managers. In the cultural change that we are embarked on, we are encouraging conversations, rather than the sense of, "It is here and it is 92.4, when it should be 92.8. For goodness’ sake, make it that." The short answer is: yes, we do have that kind of data.

The other thing we have in the system is an internal check, which is that if my offices start to do things that cause pressure elsewhere in the system, that will quickly become apparent. For example, my operational director colleagues-the ones running the benefits centres or the ones running the contact centres, for example-would quickly call out if I was just churning people off deliberately and causing them to claim or creating lots of inappropriate sanctions, because that creates work in their part of the forest, and they would immediately say, "What is going on here?"

Amyas Morse: But they would not have any idea what a standard amount of work for them would be, because everything is unique, is it not?

Neil Couling: But the jobcentres are linked into benefit centres and there is a lot of dialogue in the operational support networks looking at what is going on, such as, "Is the claimant count rising? Does that mean that we need more resources in the benefits bit to cope with the inflow of benefits?" and so on. There are quite a lot of checks and balances in the system as well.

Q94 Amyas Morse: So you-not that you are controlling it all from the centre; I am not trying to get you to say that, because I understand what you are saying-have reasonable warnings. We heard from Citizens Advice that they were worried that there were disadvantaged people who were walking into difficulty when it came to being online and so forth. They admitted that they were only seeing people who had troubles, but they were saying, "We are concerned that there is an uptick of difficulty." Would you have a way of knowing that?

Neil Couling: We bought in a new sanctions regime for JSA in October and for employment and support allowance in September, and we are watching those figures carefully to see whether some of the things that Ms Shaw was saying earlier are in fact true.

It is difficult. Ms Shaw and I have worked together for years. She helped me solve some of the problems back in 2006, so I have nothing but the highest admiration for her. Citizens Advice are seeing just a snapshot, rather than the whole caseload. We are watching this very carefully, because we understand that the reforms are quite controversial and there is a lot of public and parliamentary interest in how they are proceeding.

Amyas Morse: So there is something that you can probably share with Ms Mactaggart that would let her understand how you would do that?

Neil Couling: Yes.

Amyas Morse: I think that is what you are looking for, isn’t it?

Fiona Mactaggart: Absolutely. I am concerned about the differential impact on the most vulnerable people that is a risk of the sanctions regime. If I can be assured that you have good systems that avoid a differential impact and bear hardest on the most vulnerable, I will be satisfied.

Chair: That is another little note. I have a whole list of people who want to ask questions: Austin, Ian, Meg, Guto, Justin.

Q95 Austin Mitchell: Why do you use the fall-off from benefits as your primary performance measure, rather than getting people a job? It makes you look like the department of shakedowns, rather than the Department for Work and Pensions.

Robert Devereux: It is largely for the reasons that we have just been through. Measuring who leaves benefit and gets a job is a labour-intensive process that employers do not want to participate in and which is not, in my view, adding more value to periodic, well run sample surveys. Many things in life are measured perfectly well by surveys. Not everything is measured by administrative data. My assertion is that this is one of them.

Q96 Austin Mitchell: That looks like an excuse.

Robert Devereux: It’s not an excuse.

Q97 Austin Mitchell: If you are using the number going off benefit, it creates all kinds of perverse incentives, like shakedowns, sanctions and shoving them on to other benefits.

Robert Devereux: Again, we have just had a long conversation about whether there is any evidence that we are gratuitously using sanctions to meet our targets. There is no evidence that that is the case. If there were, this curve which we keep coming back to would have spikes in it. I am not persuaded that the off-flow target, simple as it is, is itself driving perverse performance in the way that you describe and nor is the National Audit Office.

Q98 Austin Mitchell: It would be a better check on local performance if you are giving local performance more flexibility, but let me move on. I see from figure 10 that jobcentre satisfaction levels are high in this country compared with, say, France where they just seem to tell them to "foutez le camp", which is an achievement. But how does that vary by regions of high unemployment? Could you give us the satisfaction levels in Grimsby compared with those in Cheltenham or solid south somewhere? What’s the gap?

Robert Devereux: Not off the top of my head. I am fairly sure that this is not measured at the level of Grimsby and Cheltenham. Such regional breakdown as is statistically significant, I will send to you.

Q99 Austin Mitchell: But could you give us a measure of satisfaction felt by clients in high unemployment areas compared with low unemployment areas?

Robert Devereux: If the level of sampling is such that I got a number other than a national number which I can break down, then I will break it down as much as I can for you.

Q100 Austin Mitchell: It says at paragraph 1.17 that the case loads range from a low of 118 cases per adviser in Wessex, wherever Wessex is, to a high of 213 per adviser in Birmingham and Solihull and that the Department feels that there is a link between the time advisers spend with clients and the off-flow from benefits. Surely that is just a result of the fact that you have a higher throughput in an area of high unemployment and therefore it is more difficult to get them a job. So it is a question not of the time they are spending, but of the difficulties in getting a job locally. Can you dissociate the time spent on interviews from the level of unemployment in the area?

Robert Devereux: Probably not, no. Can I make another point about this because it is not obvious to the reader? There are two sorts of staff in my jobcentres: people whom the National Audit Office and we call advisers, who typically do in-depth interviews; and we have a lot of assistant advisers who, typically, in an ordinary office, will do the fortnightly signing. So the assistant adviser’s role is: "Are you complying with your obligations? Where is your evidence of work search? Thank you very much. See you in a fortnight’s time." The adviser’s role is rather more, "Let’s see your CV. What jobs have you applied for?" That is the classic way in which it is done. We are trying hard to stop having a view from Whitehall that the right ratio is three of these to one of those for all circumstances. So in practice what we are doing is distributing financial resources around the system in relation to the stock of people on the benefit and the flow of people on the benefit because those two things are both important variables. Then we get the local manager to decide the best potential mix of assistant advisers and advisers. So this picture here is all about the adviser end. It is not about the total staff or anything else.

Q101 Austin Mitchell: I hope that means more resources coming to Grimsby. I have had complaints from people who feel that they are being treated in a rather perfunctory fashion. One chap told me that it feels like they just want to get you out of the place and off their books, rather than providing help and support to get a job; they just want to give you the bum’s rush.

Robert Devereux: They should not be doing that for all the reasons that we have been through.

Q102 Ian Swales: Can I come in on this? I think Austin has made some good points. What surprised me was that the case load per adviser does not seem to bear any relation to the type of area. My area falls into Durham and Tees valley where there are 196 cases per adviser. Similar areas with similar unemployment and job markets would be Merseyside with 155 and the Glasgow area with 152. So advisers in the area that I represent have a 30% higher case load than in areas that I would think of as quite similar. Why is that?

Neil Couling: This table reflects managerial choice rather than giving an exact understanding of the work load of individual advisers. As Mr Devereux says, in some offices, they use a plumber and mate system, where the assistant adviser does all the administrative work around the advisory interview, so the adviser can do lots more interviews a day. In other offices-in Wessex, for example, the district manager has decided to put all his resource into advisers, so his advisers have to do all the administrative tasks and do fewer advisory interviews a day.

Q103 Ian Swales: Surely, as the top people in the organisation, you must have some sort of view about what the right scheme of the organisation is. I understand local discretion, but I would have thought that it was a matter of fairly basic principle.

Neil Couling: It also has to do with the local labour market. In Wessex, which is effectively Wiltshire and Dorset, it is very difficult to recruit people at our administrative office level because of the wages compared to what is offered in the local labour market. The district manager there has decided to go after executive officer advisers and has put his money there. In other parts of the country, the wages that we pay for administrative officers are much more competitive in the local labour market, and managers there have gone to the plumber and mate system as I have described it.

Q104 Ian Swales: Merseyside has 155 cases per adviser, according to the table, and Teesside has 196. Explain to me why you think people in my area will get a similar level of service to people in Merseyside, or won’t they?

Neil Couling: What you would need to do is compare the off-flow rates rather than the inputs. You cannot conclude from the table that people in Birmingham and Solihull are getting a very different service from Wessex. What you may find in Birmingham and Solihull is that there are a lot more advisory interviews going on because the advisers just spend time in the advisory interviews and all the administration is done for them. In Wessex, there are more advisers, but they have to do more of the administrative tasks. You cannot draw that conclusion from this table.

Robert Devereux: You are picking up one variable: the number of one class and salary bracket of employee. You are not picking up the financial resources going into your area.

Q105 Ian Swales: I am not picking it up; the National Audit Office has decided that this is a useful table to have. If you are telling us that it does not mean anything, we can move on.

Robert Devereux: We are pretty much saying that. The question that you should be asking is whether we are managing, given the labour market that you represent, to serve your constituents well.

Q106 Chair: The only thing that comes out of it, Ian, is the cost. These are the top guys we are seeing. They are the more expensive guys, if I heard you right. In Wessex, they can only recruit the more expensive guys. There will be a cost per off-flow-a bit of MI-that you will have, I hope.

Neil Couling: The resource pot is equalised. It is driven by the volumes and activities. There is a table in the back of the Report setting out some of the times that we use to calculate resourcing, but the choices that the district manager then makes about what to do with that financial resource are driven partly by what he can recruit and do in that particular location and partly by his judgment about what will work best in that location.

Q107 Ian Swales: I am only using Merseyside and Teesside because I happen to represent one of them, but I think that the two areas are similar in terms of high unemployment and similar job markets. If you are telling me that they get similar financial resources per metric, as I think you are saying, then the managers in Merseyside have taken a completely different view of how to run their show than the managers in Teesside. That will not be to do with availability of administrative staff, I can tell you. Given that they have taken such a different view, do you not have any view, as manager of the Department, who has the best approach? I do not believe it should be so fundamentally different. I do not understand why it would be.

Robert Devereux: If it were genuinely, on the ground, fundamentally different you might expect that the off-flow rates between these two similar labour markets would be wildly different, but they are not.

Q108 Chair: One must be more efficient than the other or cost less. There must be something. I cannot believe the NAO has just put it in because it is a pretty table.

Max Tse: We recognise that there will be other factors than performance driving this variation. I think we were trying to encourage better understanding of what choices had been made locally, what effects that has both on cost and on performance. If you look at just off-flow, it is hard to see what is happening. If you look at just cost, you cannot see what is happening.

Q109 Chair: Is there are difference in cost?

Amyas Morse: We make this clear. In paragraphs 1.18, 1.19 and 1.20 we explain the basis of this. We do say that there are different assumptions that you use-in some cases you are using different staff mixes-so we make that clear. I know you are working on benchmarking it yourself. All we are trying to say is, even when you do clustering among apparently comparable authorities, there is quite a wide range of variation. We think it is an interesting area to understand better. We think that it is likely that-the variation looks a bit more than coincidental and it might be significant and indicative of something.

I agree with you about off-flow, but there is a question about quality of service being delivered as well as simply off-flow. I guess it is interesting just to understand that.

Q110 Chair: And value for money.

Robert Devereux: I agree with all that.

In paragraph 1.18, the NAO has given an exhaustive list of the five variables that we are already using to try to do clustering. With those five variables, we still cannot find a standard pattern. One of the variables is not, for example, the average duration on benefits. So in an area that has long-term unemployment that is not one of these variables. The only thing that I am trying to say is that real life is actually quite complicated.

I am not wholly sure there is-just for the want of another regression-a way of saying the following 93 factors actually produce happiness in all circumstances. We are trying. You can see that we have at least got as far as eight factors here. We have not yet managed, even with eight, to get to an answer. Some things in life are complicated.

If I thought for a moment that I was putting in jeopardy your constituents’ chances of getting off benefit after 13, 26 or up to 30 weeks, because they had way lower than anywhere else with higher, we would be straight back and that would be an obvious factor. But that clearly is not the case, otherwise we would have found it in the benchmarking.

Q111 Ian Swales: Okay. Can I just raise one other issue, then I will let other people come in? We heard from the earlier witness about some of the difficulties of filling in forms and so on and how people sometimes are getting referred by your people to the third sector, in the case of citizens advice bureaux. I have even had them referred to me directly by jobcentres. What are your comments about that aspect, where you have this big resource and then it seems, for some difficult cases, or cases where people are struggling with your systems, you are quite happy to push them into other areas, in some places? What is your response to that?

Neil Couling: I have heard that said, too. My experience in jobcentres is that we do not do that. Back in 2006, we were trying to move people off postal claim forms on to the telephone to claim, rather successfully, and we were concerned about some of the vulnerable people who could not cope with the telephone then. We are doing a similar thing now online. We have just put 2,000 internet access devices into our jobcentres in the last six months to help support people claim and support their job search, where they have not got the internet at home, and so forth. So we are on to some of that, but there will be some people who will want to go for what they would call independent advice, because they will see us as the person they are dealing with.

Q112 Ian Swales: If they want to, that is one thing. I am pleased to hear about this move, that sounds positive, but will your people, for example, be helping those who might have the odd language difficulty or other problems at those terminals, if necessary? Not everybody is IT-literate.

Neil Couling: Indeed. As I said, in Glasgow, Laurieston-I was there because of a Slovenian claimant-we use translation services and we try to help people through. Some cases, though, are incredibly complex, especially when you get into benefit queries. We don’t have benefit experts in all jobcentres at the level they have in the benefits centres. In some cases, people will need to go out and seek some advice elsewhere.

Q113 Ian Swales: So, if somebody gets told to go to their local library because they do not have internet access, you would regard that now as a fault in your system; they should be helped.

Robert Devereux: No, I don’t think I would actually.

Q114 Ian Swales: If they are not capable of doing it. That is the point.

Robert Devereux: If they are not capable of doing it, sure. What we have got-if you divide the number of internet access devices by the case load, you will realise there is a very small number. If people want to come to our buildings and ask staff to sit with them, we will do that. In the modern world, there are lots of other offers: one-stop shops in the library, their friends, their children, whoever it might be. Different people will get help in different ways. We are trying to ensure that we can do translation and internet access in our offices if that is what people prefer. If they prefer to take it somewhere else-

Q115 Ian Swales: As MPs, we do not want a single case where somebody fails to get the benefits that they deserve because they can’t cope with your system. That is the top and bottom of it.

Robert Devereux: Nor do I.

Q116 Meg Hillier: I want to go back momentarily to sanctions. I have certainly seen an increase in the number of people coming to see me who have been sanctioned, who would previously have been rapped on the knuckles. They are often confused. Going back to the online point Ian Swales raised, they often find it a struggle to work online and get proof that they have applied for a job. The system online goes on and they don’t know how to print off a receipt. There are some very basic things and they get very anxious.

MPs are not fools and we have to take people who come to our surgeries on trust. In the past, it was rarely applied. People would try to justify to me why they couldn’t possibly apply for a job. They didn’t get much patience. Now, people who are really trying but are not in tune with the modern way of applying online-online is the worst-are really distressed. It is often older men and people with poor English. You talked a lot about support just then that you can offer, but it is quite a lot of work for a DWP assessor with an average case load of 193 or whatever-I think it is 153 in Hackney or east London-to take someone through all of that.

For all that you say, Mr Devereux, that people can go to their children or other people, you try walking into a library, finding a hard-pressed librarian and saying, "I don’t understand how to do this." It is quite a lot of work for a librarian to do that. I don’t think there are a lot of places you can go.

Robert Devereux: To answer that, this is a really important skills set that, right across Government, we are interested in people developing. People are not going to find it easy to find work if they can’t apply online. People are not going to find it possible to transact in lots of other things with HMRC if they can’t understand online.

Q117 Meg Hillier: But they don’t.

Robert Devereux: There is a general sense that how you make adequate provision to give assisted support for digital is a Government aspiration, not just a DWP one.

Q118 Meg Hillier: Good. I am glad it is a Government aspiration. So now, what are you doing?

Robert Devereux: Within our offices, we are trying, within the resource levels that the Treasury gives us, to ensure that we can do that. But this is not an infinite elastic amount of resource. The advisers are trying to make judgments about where they can best support people. I have seen with my own eyes, as Neil has, people sitting down and helping people talking through this, doing it in group sessions, being innovative in different ways.

The whole point about trying to have empowered local staff is that they are also very good at making connections around the place. The other day I was in Glasgow in Shettleston road, where the staff have created a youth hub for young people and they have brought in lots of different providers who bring different things to the party, all of whom say that their own performance in terms of getting people into work has gone up as a consequence of working together.

They have also been very positive about being able to move people around between different providers and give different opportunities. I don’t want it to come across as though there is only one answer to this question, which is that these people do it. Different staff are thinking differently and innovatively about different ways to support people.

Q119 Meg Hillier: I don’t doubt that. There are good people. The Peabody Trust has a jobcentre on the Pembury estate in my constituency that does some very good work. I think sometimes it is embarrassment. I come back to older men, as I seem to have seen a run of them recently. They are embarrassed to say that they don’t know quite how to do it. They think they have done it right, and then they cannot get the relevant bit of paper that proves they have done something to show to the jobcentre adviser. Some of them are used to getting jobs that do not require them to apply online. In the old days, they did not have to do that.

Neil Couling: Twenty-five per cent of job vacancies at the moment are only available online, and that is growing. So we have to help people make that change.

Q120 Meg Hillier: And 75%-

Neil Couling: Yes, but 75% is a shrinking number. With universal job match, we have been helping people to register. It is the new internet-based vacancy system that we are running now. You asked a question about how I can improve my job search if I am not very good at using the internet. Well, there is a functionality in the system that if the claimant grants us access to their job search-so not their personal details-it is visible to the advisers. So when the person comes to sign on, the adviser can click on and say, "Yes, I can see you have applied for four jobs."

Q121 Meg Hillier: That is if they do it through your portal.

Neil Couling: That is if they do it through our portal. The technology will help with some of these responsibilities. It is a question of supporting individuals through to use that and to get familiar with it.

Q122 Meg Hillier: That is good news. What I am concerned about is the fact that the first thing someone who is clearly confused and worried about the system faces is a sanction rather than being given support. It feels like that. I am sure that the cases my colleagues are receiving show that as well.

Robert Devereux: Certainly in respect of online, the challenge was that they were unable to demonstrate that they had done their online activity properly because they could not get a bit of paper out. If that is the nature of their problem, but they are diligently searching online, our staff will be able to see that if they are using universal job match. It is a particularly good regime for that, because it takes a lot of the bits of paper out of the system and you can see exactly what is going on. Better than that, you can see that they were successful in getting to interview.

Q123 Meg Hillier: To be clear, that applies only to jobs advertised at the jobcentre, and not other jobs.

Neil Couling: There is a notes facility that allows you to record things such as, "I went round six shops and asked whether they had any jobs." Yes, it records what is on universal jobs match, but it also allows you to record when you have been on other sites.

Q124 Meg Hillier: It sounds great, and I am not knocking any of that. I think that that is all good news. But I do know people who have been around shops saying, "Here’s my CV. I would like work" and that does not count and they have been sanctioned. Perhaps I will raise some individual cases with you.

Robert Devereux: That should not be the case. If you have examples of people who have a perfectly diligent job search that is not being counted as a job search, let us know.

Q125 Meg Hillier: I will certainly do so. I have a really important issue about courses, and it relates possibly to the Work programme and to jobcentre advisers as well. A number of people from my local community college have been on a course that was getting them to a qualification or getting them to be more employable and they were told to come off it or they would be sanctioned because they were not available for work. That happened to somebody a week before they finished their course. That seems to be completely counter-intuitive. I would hope that that has never been the intention of the DWP. If it is, tell us now, but otherwise why is that happening so often? I have a raft of cases which I will certainly share with you, because it is a matter of real concern.

Neil Couling: I probably write back to Members of Parliament on some of those cases quite regularly as well. It is a very complex area of social security law-how the social security system interacts with the education system. Social security is not there to provide support for people studying.

Q126 Meg Hillier: But the Work programme is.

Neil Couling: In terms of how the rules work, availability and actively seeking tend to trump any of those other aspects, and that has long been the case. There was a concern in the 1980s or early 1990s that the social security system was effectively being used to support people in education when there were other forms of support. In those days, it was grants and now it is loans. Each case is tricky to advise on on behalf of my folk, and to make decisions on.

Q127 Meg Hillier: You talked about empowering staff at the front line and giving them discretion. If someone is within a week of finishing their course, it seems bonkers to take them off. Also, many students say to me, "I study, but I still want to work; I still have time to work up to 16 hours or even more a week," because there is a lot of shift work available for supermarket workers and so on. You can do both.

Robert Devereux: All my staff are trying to do is to apply the law as passed by Parliament. If Parliament asks that as a condition of JSA, people must be available for and actively seeking work and there are definitions of what both of those mean, it should not be the case-

Q128 Meg Hillier: But if they are taken off the course because the Work programme is putting them on another course, that is just bonkers, isn’t it? That is happening.

Robert Devereux: You are asking me about the JSA regime. All Neil and I are trying to say is that-

Q129 Meg Hillier: But people on the Work programme are on JSA, too.

Robert Devereux: If you want to give me a pot of cash from Parliament to say, "Do good things with the people who come through your offices," I will happily do that, but that is not the way you have set it up.

Meg Hillier: No, but if you are on the Work programme, you are trying to get work and the Work programme can put you on courses. They are taking people off existing courses that are actually in their line of specialism or that they are nearly finishing and putting them on something else, so they are on a course and in receipt of benefit, but not on the course that is actually going to help them. It just seems bonkers.

Chair: I am going to move us on, because this is the sort of flexibility you would hope local jobcentres-[Interruption.] I think probably you should correspond between the two on the particular issue. I am just conscious that we have to get a move on.

Q130 Guto Bebb: My questions relate to the way in which the jobcentre works with the Work programme. On the sanctions, to start off with, one of you made the comment that the sanctions often depended on the extent of the support being offered to the clients. I am just wondering. Let’s say that somebody has been referred to the Work programme. Is there a higher level of sanctions for people on the Work programme, as compared with those who are not yet on the Work programme?

Robert Devereux: It was me who made the point. I wasn’t quite saying that it depended on the extent of help. What I was saying was that where our front-line staff have put a lot of energy and innovation into supported offers-

Guto Bebb: I am not being critical, by the way.

Robert Devereux: I have noticed that they are then much more confident in the difficult act, on behalf of society, of sanctioning somebody. I am not saying that because there’s more offer, there’s a higher rate of sanction. I am observing that it is not surprising; it is human nature that if you are confident that you have really pushed the boat out and given people lots of opportunities, you are a bit more confident about doing the sanction thereafter. That was the point.

Q131 Guto Bebb: The point I am making is that Work programme providers will often feel that that is the case in terms of the support that they offer. Is there any differential between Work programme providers-

Neil Couling: Broadly-and this comes out in the Report as well-there is a link between frequency of sanction and duration of claim. Typically, at the start of a claim, there is a very low level of sanctioning, and that grows; it is not intuitively the wrong way round, I don’t think. That is a rough rule of thumb. If you are thinking about the Work programme, they are already getting people who have been on benefit for at least a year, so you might see in the data a higher sanctioning rate than if you look at the whole JSA population.

Q132 Guto Bebb: Is that the case, then, that there is a higher sanctioning rate?

Neil Couling: I do not think it is that marked. It would depend on which cohort of jobseekers you compared it against.

Q133 Guto Bebb: The point I was going to make was that, having visited several Work programme providers, the feeling I had was that that would be the final resort in any situation. I was just wondering, out of interest.

Robert Devereux: First, it is not wholly in the gift of the Work programme to sanction somebody. As we have discussed previously, sanction decisions are made by members of staff trained in understanding whether there is good evidence, so it is true that the Work programme provider may recommend somebody for a sanction-just like one of my members of staff-but unless there is evidence in front of that decision maker, they will decide that actually that does not work, either because the evidence is not there or because there is a very good reason-"My mother died yesterday"; "good cause" reasons. The check and the balance in the system is that I have not given either Neil’s staff or Work programme staff the facility to make a sanction, only to recommend such a thing to a third party, on behalf of society, who tries to make a reasonable judgment. They are well trained in doing that.

Q134 Guto Bebb: Okay, that’s fine. The second point is that figure 12 in the Report seems to imply that there is possibly an element of parking occurring with some claimants. Whether that is the case or not, I wouldn’t mind a comment on that. It does reflect some of the comments I have had from Work programme providers, who claim that they are having increasingly difficult clients to deal with. I just wonder whether you would care to comment on that issue-whether that figure is indicating parking and whether that reflects the fact that Work programme providers are complaining of having difficult clients to deal with.

Robert Devereux: I do not think this figure does demonstrate parking. The best that the NAO could come up with was the risk that it might be. Let’s just explain what is going on here. This is a chart trying to record what proportion of the case load of long-term jobseekers are actually being, as it were, referred to a third party. It is not trying to measure the amount of effort that my own staff are putting into this. It is only to do with whether there was a third-party opportunity there.

If you take yourself back to 2009-10, at the height of the new deal, the future jobs fund and the six months offer, there was a whole bunch of prescription that basically said, "When you get to this place, you shall do this." As a consequence, the proportion of people who seemingly were not being sent to them was very low. Wind yourself on to 2011-12, when some of those offers have disappeared precisely because we want to give jobcentres more flexibility to decide what is in the best interests of somebody, rather than treating them just as a statistic, actually they are making different choices.

To get this into perspective, this chart talks about having sampled 10,000 jobseekers. Those 19% in 2011-12 boiled down to 43 people. By the time you have taken the long-term out, and taken everything else out, these are quite small numbers. My own view is that this is picking up a tiny piece of the arrangement-

Q135 Chair: Do you agree with that, Max?

Max Tse: I agree it is a relatively small sample and there may be other things going on, so we state that in paragraph 2.7. But we did try to check whether there was more personal adviser activity for the group that were not seeing referrals, and we did not see any difference between them and the other groups.

Robert Devereux: So the question you need to ask yourself is whether or not-what about the quality and appropriateness of the referral? It is easy, particularly in the world where looking busy is the way to be, to just refer everybody, but whether that helps anybody-I answered as many letters as you have from people saying, "I was sent on this course. It was ridiculous. I’ve already got a CV. I don’t want to be in that space."

The other thing comes back to the bottom line. Where is the evidence that that significant change, apparently, as has been referred to, is showing up in these data? I am not sure it is, actually.

Q136 Chair: Your own research and evaluation of the Jobcentre Plus offer-the DWP’s own evaluation-suggested ESA claimants do not seem to discuss or receive support to the same extent as jobseeker’s allowance claimants, with that being particularly a concern for those who are currently looking for work around 16% of the year, said claimants surveyed. You go on to say that, although it is true that the majority of ESA claimants-70%-discuss the possibility of working in the future, this means that a substantial minority of 30% did not. Similarly, nearly half of all ESA claimants did not discuss what steps they could begin to take to find work in future in the new work-focused interview. All I am doing is reporting to you your own research. In addition, ESA claimants were significantly more likely to report that they left their initial meeting with their adviser without an appointment for their next meeting.

There are three bits. I am sure your Mr Couling, in particular, will be familiar with your own research, which suggests that ESA claimants are getting a worse deal from Jobcentre Plus.

Robert Devereux: So this is a Report wholly about jobseeker’s allowance. The data in here-

Q137 Chair: This is about Jobcentre Plus.

Robert Devereux: Just to be clear, everything you read out about ESA, which is indeed our own research, is about claiming ESA. All these pictures we are looking at are about JSA, so I will-

Q138 Chair: It does not matter. The point is the same. The ones who are on JSA in the longer term will tend to be the same people-those who have come off incapacity benefits and gone on to jobseeker’s allowance-which is fine. But what is happening is that those more vulnerable and distant from the labour market in two bits of research-it seems your own evidence is suggesting they are having less support. You are parking them.

Neil Couling: "Parking them" is an emotive term, and I do not think we are. Prior to ESA we were not doing anything with incapacity benefit claimants in jobcentres.

Q139 Chair: You are doing less with them than you are with people who are closer to the labour market.

Neil Couling: We are doing less-

Robert Devereux: They are close to the labour market; they are on JSA. The whole point of the change of ESA is to make sure that we are trying to find out who should be subject to the full JSA and if they do not-

Q140 Chair: Either your own evidence is right or it is wrong.

Neil Couling: I was just coming on to explain that we do less with ESA customers than we do with jobseeker’s allowance customers. That is one of the reasons why we are reassessing incapacity benefit caseload. We are encouraged when we put somebody on to JSA, even though they personally, themselves, might not be encouraged, because we know that is a good way of getting people into work. We know the JSA regime is successful.

In addition to that, we have also changed the referrals into the Work programme, so that ESA customers go much earlier into the Work programme now, and effectively access extra support there. The fact that jobcentres are not working with ESA customers, which is what that piece of research said, was a policy response-they were moved over into the Work programme, which is where they are now being dealt with.

Q141 Chair: We did not see that in the early stats, so we look forward to seeing it in the later stats.

Neil Couling: No, it was a change we made some eight or nine months into the operation of the Work programme.

Chair: It might have been picked up in the early stats. Sorry Guto, I interrupted.

Q142 Guto Bebb: Specifically on the ESA, are they being compulsorily or voluntarily referred?

Neil Couling: They are being compulsorily referred at three months now, in the main-after they have had their work capability assessment decision.

Chair: We look forward to seeing that in the stats.

Q143 Justin Tomlinson: A couple of quick things. First, some real credit to Jobcentre Plus. My constituency heard recently the announcement of Honda job losses, and Jobcentre Plus has been fantastic in helping proactively to co-ordinate support ahead of those workers entering the unemployment market. Also, my wife volunteers helping to run a job club and she was also full of praise about the universal job match. People are now able to put their CVs online so that employers can proactively contact them. Some of the people for whom they found it difficult to find jobs have suddenly got hold of things out of the blue. Also, the inclusion of alternative listings, such as Monster and others, all in one place has made a huge difference. However, she said that the big challenge is that nobody can ever remember their own password to get on to the gateway system and that that takes up a big chunk of their time.

One other point, referring back to figure 8, I completely understand that that obviously shows outcomes. During this meeting I have heard from my office. My Swindon constituency has just transferred from Wessex, which is at 118, to Thames Valley. We have already found a different approach to things. The co-ordinators we are working with are saying that they do not have the right tools, cannot access the same information and that the authority to provide help is not available. I understand about the different ways of doing different things, but what emphasis is being put on sharing best practice?

Neil Couling: I looked at how freedoms and flexibilities were running out-I think it was Mr Bacon who said that we had had a dragon’s den, and I wound the dragon’s den up. Basically, a very competitive regime was going on and people were not sharing. So I set up a whole series of mechanisms to allow and encourage sharing. A lot of local sharing is going on, but there is also now quite a lot of national sharing. Some of the jobcentre performance in London at the moment is extraordinarily strong. I notice that Mr Swales has gone now, but there are quite a lot of people coming down from the north-east, for example, to look at what people are doing in London and learn from it.

Q144 Justin Tomlinson: Crucially, what about organisations? We as Members of Parliament might deal with this, as might job clubs or whoever. How can they put in what they see as best practice, because when you speak to the people who are doing it, they are all going to say, "We’re doing it the right way; ours is the best way and we are happy to share it."?

Neil Couling: I would encourage a warm and close relationship with the jobcentre district managers. They ought to be very open to building those kind of relationships, because finding somebody a job is not just something that a jobcentre does-it is about being in the community and working in that community, and understanding it. I would just encourage that to happen.

Q145 Chair: I am just going to go through some issues that I do not think we have covered. One of the things that your own evidence says is that the more time you spend with people, the more likely they are to find a job-that is on page 18, at paragraph 1.17. Yet elsewhere in the Report, it says that less time is being spent, for example, on work-focused interviews. Fewer are being done-it has gone from one in four to one in five. So you appear to know what works-individuals spending more time with the adviser-yet probably one of your main vehicles for doing that is your work-focused interviews and you are doing less of them. Can you explain that?

Robert Devereux: So if I am correctly guessing how you are doing the sums, your one in five is coming from figure 4.

Q146 Chair: I looked at page 8, paragraph 9, where it says that work-focused interviews fell from one in four to one in five. Yet your analysis on page 18, paragraph 1.17, says that there is "a link between the time advisers spend with clients and off-flow from benefits."

Robert Devereux: So if we have a peek at figure 4, which is where the data sits for that bit of the summary-I need to explain something about how the case side matures when you make a big shock to it. In the picture in figure 2, you can see that the National Audit Office has identified that the claimant count rose very quickly in the back end of 2008-09. The nature of the interventions that we do with people typically leaves them pretty much to get on with it in the first few weeks. We do progressively more with them the longer they are on benefit.

What you then find, effectively, is that if you change the nature of the make-up of the caseload very quickly, as we did in 2008, you would expect to see something reducing in the proportions over 2009-10 and then coming back in 2010-11. If you try the same thing a bit further down in the sanctions paragraph, which is 1.11, it appears to suggest that sanctions fell off a cliff and have gone back again. Actually, however, if you track through when it is more likely that you would be sanctioned because of the length of time on benefit, you find that that is an entirely standard pattern.

I am not sure I am with you that we have somehow or other wilfully taken time away from things that work. The nature of the caseload was moving very sharply in those three years. As soon as you go back to 2010-11, we are back to pretty much the same ratios that we had in the previous staple period. It is a rather complex answer, but that is the answer.

Q147 Chair: Right. So why does the Report say otherwise?

Robert Devereux: The Report does not say otherwise. You have brought together a statement that I am not reneging on, which is that more adviser time, all things being equal, is probably better.

Q148 Chair: Okay. Good. Let me ask another question. Jobcentre Plus is no longer an agency and comes under your direct control. I do not know what benefits were intended, but have you achieved the benefits that you wanted?

Robert Devereux: Yes, I have.

Q149 Chair: What are they?

Robert Devereux: There are two things that have happened. One is that there is an overhead in having an internal organisation inside another organisation. I have non-executive directors, I have sets of accounts and I have the attitudes of people in the agency versus the rest of the Department, all of which I might be indifferent to in a world where I had plenty of cash. In a world where I have to take 30% out of my staffing over the Parliament, that becomes an issue.

More importantly in my view, as Mr Couling has made plain, there are so many connections in different parts of our business that thinking I can hermetically seal the working aids from what is going on with disability or the world of parents splitting up and trying to look after their children does not get me to a cultural aspect in the Department, where people are looking at what we are managing to achieve as a Department. Last time I looked, all these people work for me. I would quite like them to be thinking, "How is this working for the claimant? How does this work for the taxpayer?" That is the only really interesting question. There are days when, if you construct agencies, you can get into a not good place.

Let me be clear: this answer is in respect of my Department as it stands now. I used to run the Department for Transport, and the Driving Standards Agency is a perfectly good agency. Neither the permanent secretary nor the Secretary of State spent a lot of time fretting about the DSA. It could be perfectly well run as an agency with a board and a chief executive. That seems fine. When it comes to something that is absolutely mission critical to the Secretary of State and me-how the labour market functions-I am not persuaded that I get the right return in this austerity climate by setting it up as an agency. I have saved cost, but also liberated a degree of working across boundaries that previously was not there.

Q150 Mr Bacon: Just briefly, in parentheses, is the same now true of the Child Maintenance and Enforcement Commission? Is that now fully back under the Secretary of State directly?

Robert Devereux: The Child Maintenance and Enforcement Commission used to be a Crown non-departmental public body. It is now part of the Department.

Q151 Mr Bacon: So it never actually became an executive agency? There was an announcement that it was going to, but it never did.

Robert Devereux: Sorry, I am stating a difference. I think it is technically a Crown NDPB. It is a slightly odd category thing. It was definitely out there away from the Department.

Neil Couling: It was an agency, and then it became a Crown NDPB.

Robert Devereux: Now it is back in the Department. At the moment, there is no substantial difference between the staff in Falkirk running the child maintenance system and the staff in the Swansea pensions centre.

Q152 Mr Bacon: They are directly under the DWP.

Robert Devereux: They are directly under the DWP.

Q153 Chair: I am conscious about time. We have talked a little about employment and support allowance and the work that you are doing with disabled people. I was quite surprised to see in answer to a PQ that there are only 522 disability employment advisers across the whole of Jobcentre Plus. That is a recent one you will be familiar with. No doubt you have prepared for it. That means that there is not even one per Jobcentre Plus, per office.

Robert Devereux: Yes.

Neil Couling: As we took the decision to move ESA claimants over to the Work programme, we made consequential changes to the number of people we have got working with that kind of caseload. Effectively, our work with that group shrank and is now being done over in the Work programme.

Q154 Chair: So the first three months they get nothing. For the first three months they are on ESA or whatever it is. The first three months they will get nothing because you haven’t got anybody. That is not good for people who are on a disability benefit. The quicker you can get it, the better, as always.

Neil Couling: Again, it depends which question you ask, as always. The disability employment advisers are not the totality of people we have got in jobcentres working with ESA customers.

Q155 Chair: No, they never were, but not having even one per jobcentre seems a bit harsh, to put it mildly. There are 522, and 740 jobcentres. I hear that you have taken a decision. I am just making the observation that, knowing the sort of people who go on to ESA, if you don’t capture them early-

Neil Couling: There are some very small jobcentres that are probably non-viable in terms of their sizes but we keep them open because of their geographical location. In some of those it would not make economic sense to have a full-time disability employment adviser.

Q156 Chair: Are there 220 like that?

Neil Couling: There are some very small jobcentres.

Q157 Chair: 220 out of 740? That is the number. If you have 522 DEAs and 740 jobcentres at the moment, that means 220 have not got a DEA.

Robert Devereux: You are quoting a number I have not seen; is it full-time equivalent or the heads?

Q158 Chair: It is a PQ answer, which I do not have. It is 522 disability employment advisers.

Robert Devereux: Okay. If it was full-time equivalent there could be more heads.

Q159 Chair: Are there? You should know that.

Robert Devereux: I don’t recall that.

Q160 Chair: Well, are there? Mr Couling should know; he is running it.

Neil Couling: I do not know the answer.

Max Tse: Figure 1 has 560 FTEs doing employment and support allowance and incapacity benefit. That is slightly different but it is probably for a different year.

Robert Devereux: 560 FTEs-full-time equivalents.

Q161 Chair: But there are 740 jobcentres.

Robert Devereux: That could well be a presence in most. It is definitely not the case that it is present in all of them. I have been to Malvern, where there are three or four people, and I do not think there is one there. I don’t think it is as pronounced as you are implying by saying-

Q162 Chair: Okay, maybe you could do us a note on that.

Robert Devereux: We could certainly do you a note on that.

Q163 Chair: Another question is about a quarter of your expenditure going on offices at the moment. I have no idea what it is in banks; we haven’t been able to check it. It seemed quite a lot to me. You are telling me that some of the offices are supposed to be the front line, particularly as you are expecting more and more people to come to you online. That seems a lot of the money-I don’t know whether it is, but it is just a really interesting question. About 25% of your expenditure currently is just on actually keeping the offices going.

Robert Devereux: Okay, let’s think about the nature of the business here. We are trying to interact with people’s behaviour to ensure that they are continually fulfilling obligations.

Q164 Chair: I am just asking whether you think that is a sensible figure.

Robert Devereux: It seems to me that if I wanted to persuade you to be in work and checking what you are doing and supporting you, that is almost certainly done well face to face. That is why we have a distributed network. Whenever we seek to say to any MP that we are going to close the jobcentre in their area because we think we can do it remotely, the answer is always no, because actually it is a really important part-

Q165 Chair: Do you think it is a sensible figure, Mr Devereux?

Robert Devereux: Yes.

Q166 Chair: Okay. We will check it against what banks spend.

Robert Devereux: And why would you do that? What is the nature of the banking business that is so similar to the behavioural-

Q167 Chair: It is a similar one that does it face to face; that is what we thought.

Robert Devereux: I have a perfectly good bank account and go nowhere near the bank. But I am not trying to get work.

Q168 Chair: You are trying to encourage 80% of your people, for example, to apply for benefits online.

Robert Devereux: Yes, to apply online. When it comes to eyeballing members of the public to make sure that I am confident they are doing the right thing and they feel that they can get the right support, that is not the same as applying online. By all means, let us take cost out of things to do with paper and telephony that are not needed for the claims process. But when it comes to behaviour change, all the evidence shows that this labour market works better than many others around the world because we have an active labour market, and we have good-quality staff doing fortnightly job signing. Why is it that the British labour market performs well relative to others, and why do other people come and copy us? It is because we do it well.

Q169 Chair: We don’t know because you don’t monitor it. Can I just ask one thing that really concerned me? On page 32, paragraph 2.24-I think this is the right reference-it says that one third of claims are not pushed. A basic function in Jobcentre Plus is to be efficient about processing claims and to make sure you have the right info to go in. According to the Report, there is an extra cost to the Department of 450 full-time equivalents because of the inefficiencies in that bit of the system. That is a pretty incredible figure if it is true.

Neil Couling: I need to tell you the story of that because you might draw the wrong conclusion. Ms Shaw can testify that the problem we sorted out in 2006 was exactly this one, and what to do. Essentially, as part of the process, people in jobcentres receive the evidence on a claim and verify what is in front of them. In 2006, the rule was that if you had not verified absolutely everything, you should hold on to the claim and keep talking to the claimant until the verification is done. What did that mean? Claim times doubled, and we had filing cabinets full of claims in offices and complaints from Members of Parliament and others about the situation. As the then benefits director running the back-office operations, I turned that on its head and said that I would turn a blind eye to the fact that some cases will come through to me as not complete in terms of evidence, and I will process from the benefits centres because they have the experts who can interact with people and explain exactly what piece of information will do.

What happened? As Ms Shaw will evidence, the claim times fell not just back to where the targets were, but by another third. Customer satisfaction with us improved, and we pushed cases through. This issue is about where you want your inefficiency. Do you want it in the jobcentres or in the benefit centres? The best place to put it is in the benefit centres because they can put the claims right, and can put the claims into payment. If you put it in the jobcentres and you put pressure on them to try to get absolutely everything right, they will try their very best but it will slow the process up and put cost into the system because the claimant will ring up the benefit centre or come into the jobcentre and say, "What is happening?" I made a pragmatic decision in 2006. I think it was the right one then and I think it is the right one now. The National Audit Office is right: it is 450 extra people. But that number would be a lot higher if I tried to do it by sticking it back into the jobcentres.

Chair: Okay. We are changing our tack a bit, and then I have a couple of final questions.

Q170 Mr Bacon: I have just a couple of questions relating to the previous Report on preventing fraud and contracted employment programmes. When the Department gave its response, we had a recommendation about whether there was a test for whether someone is fit and proper. The Department’s answer was that there is no "fit and proper" contractor test; there is no recognised statutory test in the context of the procurement rules. I have one or two quick questions about that. Plainly, even if there is not a statutory "fit and proper" test, that does not prevent you from providing transparent criteria about how you assess contractors. Why are you not taking that forward?

Robert Devereux: I think the answer we have given you is that in order to be compliant with the public procurement rules, we have to stick to whatever they say you can do. One of the things they do not invite you to do is go around deciding whether somebody is fit and proper.

Q171 Mr Bacon: But if it should turn out that they are not fit and proper-

Robert Devereux: Sorry, I hadn’t finished the sentence. What we must do is make sure that we make the best possible decisions, consistent with the law, about how to assess people. One thing-just to make sure we are clear here-is how you would establish such a thing on the proper basis. If I decide that I do not want to give the business to Richard Bacon Inc. because I just don’t like him and I decide he is not fit and proper, that is not going to work very well. I need to make sure that there is some sort of absolute test. To my understanding, what the note is saying is that there is not a way of doing that that will stand up in procurement law as a test. That is not the same as saying that we should not be really clear about what it is that we are trying to find, and what sort of skill sets and evidence we are trying to produce here. As I understand it, you cannot simply introduce a "fit and proper" test as part of the procurement.

Q172 Mr Bacon: If you are saying that there are indeed characteristics that you would look for, would you support the creation of an industry-wide code, which sets out what the characteristics are for being fit and proper, and using that to assess contractors? It would surely help.

Robert Devereux: So long as it was consistent with the law for procurement, which I am not an expert in.

Q173 Chair: Whose law is this?

Robert Devereux: The law of the land.

Q174 Chair: Ours or European?

Robert Devereux: It is ours.

Q175 Chair: We cannot, in contract law, decide-it belies common sense a little bit.

Robert Devereux: If it belies common sense, that may be the effect. As to whether it is true, if I have written it down here, my procurement people have clearly looked it up and checked what is possible.

Q176 Chair: Sometimes we find that procurement people go a little bit OTT on what they can and cannot do.

Robert Devereux: I have brought it with me. It is a very thin sentence. It says, "A fit and proper contractor test is not a recognised statutory test in the context of the procurement rules." It goes on for three more sentences to explain that even more.

Q177 Chair: Does it say we cannot put it in? Just think about it logically. Take it away from the law, and think with a bit of common sense. It makes sense, particularly given the stir there has been around this whole thing, to have some definition from you as to what level of poor practice, or intentional or unintentional fraud, you will tolerate as the purchaser of contracts in the Work programme. It would make sense to have that, and it would also make sense for that to be transparent. It surprises me that that rather common-sense approach is not allowed in the law.

Robert Devereux: It may be because it is more difficult to establish the test you have just loosely described as something that can apply in practice.

Q178 Chair: Why?

Robert Devereux: Because your view of what past fraudulent practice is, or of past service standards not being appropriate, may not be somebody else’s.

Q179 Chair: No, it may not be, but that does not stop the Department and the Government having a view, and making that transparent, so that perhaps we round this table can understand at what level you will start feeling it is inappropriate.

Robert Devereux: We have made it perfectly clear, on what must be the fourth time we have been round this buoy, that we are not in any way permitting anybody to commit fraud on our contracts. We chase after it.

Q180 Chair: In your letter to me, you talk about it being only a small amount, so you are taking a view that a small amount of fraud and mismanagement is acceptable. I want to know at what level-we had this in a hearing last Thursday; it was the same thing-it becomes unacceptable. Is it £50,000? Is it £5 million? Is it £50 million? £500,000? I do not know. What is the level at which you will say to me, "Actually, this is not acceptable"?

Robert Devereux: As a condition of bidding for any further Government contract? The only point of having it in these tests would be to remove you from a competition.

Q181 Chair: Or stop an existing contract. What level? What level becomes unacceptable for HM Government?

Robert Devereux: To stop an existing contract, it would be a matter of looking at the level of fraud. I am afraid that that would be a matter of judgment, and I am not going to give you a simple answer that says, "If it is less than that, I am tolerating it." My basic attitude is that I am not tolerating this, but I do not want to set a level and say, "So long as it’s less than 5% I will probably let you have the contract," because what they have done may actually be quite egregious.

Q182 Chair: How bad do they have to be before you think you have to intervene? I want to have some understanding of it. I do not think that is unreasonable. You are responsible for dispensing public money. Through all this, we have discovered some fraud. You think it is trivial; at what level does it cease to be trivial?

Robert Devereux: I am afraid I cannot answer that question. All I can do is deal with the allegations you put in front of me. We have looked at those. They do not seem to me to be material, in that "This contract must be taken off that person." That must be the judgment we are reaching; otherwise we would have taken it off them.

Q183Chair: It is not unreasonable to say to you that you have got to be rather more transparent about the level at which they cease to be trivial.

Robert Devereux: I worry that the risk in doing so is that it condones a level of fraud. You are basically asking me to say, "Beyond this level, I will take the contract off you; below that, it is okay." Surely it should depend on the circumstances and the nature of the contract. I am not trying to be clever, but I worry that a simple black-and-white rule will turn out to have something black in it, which you do not want any more than I do.

Q184 Mr Bacon: Would that be the issue if you were to have a definition? You say in your response that there is no definition of systemic fraud, and that each case needs to be considered on its merits. Are you saying the same about that? Were you to have a definition, would people be able to find a way of doing things that were by definition not systemic fraud?

Robert Devereux: There is a risk in codifying things that it makes them sound simple. "Surely, Mr Devereux, you must have a definition of fraud." Actually, it is not so obvious that I should. Surely we should both-

Q185 Mr Bacon: Are you saying the same about a definition of systemic fraud?

Robert Devereux: I think it is possible to define systemic fraud without it becoming a trigger point. I am not prepared to say that systemic fraud is x%, and if it is less than x% it is not systemic, because I would be straight back into the problem. Please do not get me wrong. If you have a great way of getting through this, I would love to know, but I do not think the answer is, "Surely, Mr Devereux, you must have a definition," because I suspect all that will do is condone an acceptable level, which I do not condone.

Q186 Mr Bacon: One of the other recommendations we made, because we concluded that there was no obvious route through which clients, contractors, employees, MPs or members of the public could raise issues of concern relating to fraud or poor service, was: "The Department should publicise its arrangements to enable whistleblowers to make complaints and capture and analyse information about complaints made about providers." What new work have you done to publicise whistleblowing and complaints procedures?

Robert Devereux: I may have to come back on that, because I do not know the answer to that question.

Q187 Mr Bacon: Is it likely that the answer is none?

Robert Devereux: The answer we gave you was that we believed we had already made all employment rights and service framework contractors provide that.

Q188 Mr Bacon: Your response basically restates your existing procedures, does it not?

Robert Devereux: Yes, it does.

Q189 Mr Bacon: What I really want to know is whether, since we recommended you publicise your arrangements to enable whistleblowers to make complaints, any new work had been done on publicising whistleblowing complaints procedures.

Robert Devereux: I suspect the answer is no, but I will check for you.

Q190Chair: One very final question. There was a report in The Guardian last week on universal credit. Have you suspended the contract with Accenture?

Robert Devereux: No.

Q191 Mr Bacon: Is testing starting next month? I have been asking you this question for about 18 months.

Robert Devereux: Testing is already under way.

Q192 Mr Bacon: It has started already. It started early, did it?

Robert Devereux: Testing takes a long time.

Q193 Mr Bacon: Yes, but wasn’t it originally scheduled to start in April 2013 and go on for six months?

Robert Devereux: You may be distinguishing between what is necessary for what we described as the start of a national roll-out from October, and the start of the pathfinders. What I have just answered is that in order for the pathfinder to start, there has to be some IT present. That is already in testing now.

Q194 Mr Bacon: So it is all going swimmingly, is it?

Robert Devereux: It is all going swimmingly. It is a hard project.

Q195Chair: Have you in any way changed your relationship with Accenture on universal credit?

Robert Devereux: No. Accenture, HP and IBM are working with us to complete the pathfinder and take universal credit-

Q196Chair: Have you changed your relationship?

Robert Devereux: No.

Q197 Mr Bacon: Has the scope changed?

Robert Devereux: The pathfinder is still the pathfinder, and we are still working on the plans for what will be rolled out, because that is what we are doing now.

Q198 Mr Bacon: Has the scope of what will be rolled out changed?

Robert Devereux: No. Universal credit is a system that Parliament has already legislated for, and we intend to implement it as legislated.

Q199 Mr Bacon: Has the timetable for what will be rolled out changed?

Robert Devereux: No. If we had made changes to what we are going to do, we would have come back and told you.

Chair: Thank you.

[1] Note by witness: My answer should have said that ‘In the first quarter [not the first two quarters] ... The rise in the second quarter was 25% (not 35% as written) and 45% in the third quarter.

Prepared 18th June 2013