To be published a s HC 479-i

House of COMMONS



Work and Pensions Committee

The role of Jobcentre Plus in the reformed welfare system

Wednesday 26 June 2013

Nilufer Rahim, Matthew Oakley, Tony Wilson and Adam Sharples

Evidence heard in Public Questions 1-65


1. This is a corrected transcript of evidence taken in public and reported to the House. The transcript has been placed on the internet on the authority of the Committee, and copies have been made available by the Vote Office for the use of Members and others.

2. The transcript is an approved formal record of these proceedings. It will be printed in due course.

Oral Evidence

Taken before the Work and Pensions Committee

on Wednesday 26 June 2013

Members present:

Dame Anne Begg (Chair)

Jane Ellison

Graham Evans

Mike Freer

Sheila Gilmore

Glenda Jackson

Stephen Lloyd

Nigel Mills

Anne Marie Morris

Teresa Pearce


Examination of Witnesses

Witnesses: Nilufer Rahim, Senior Researcher, NatCen Social Research, Matthew Oakley, Head of Economics and Social Policy, Policy Exchange, Tony Wilson, Policy Director, Centre for Economic and Social Inclusion, and Adam Sharples CB, Non-Executive Chairman, Ixion Holdings Ltd and former Director General, Employment Group, DWP, gave evidence.

Q1 Chair: Thanks very much for coming along this morning. I am sorry you have had to wait for some time. We are also struggling a wee bit with the new technology. The Committee has just gone paperless, and it is not quite as easy as we thought, which may be a portent of the way that Jobcentre Plus is going as well, in terms of digital by default. We are finding it quite hard to be digital by default. Anyway, thank you very much for coming along. This is our first evidence session in our inquiry into the effectiveness of Jobcentre Plus and its role in the future. I know that, between you all, you have a huge breadth of experience. Can I start by asking you-starting with you, Matthew-to introduce yourselves for the record, please?

Matthew Oakley: Good morning. I am Matthew Oakley, Head of Economics and Social Policy at the think-tank Policy Exchange.

Tony Wilson: My name is Tony Wilson. I am the Policy Director at the Centre for Economic and Social Inclusion.

Nilufer Rahim: I am Nilufer Rahim. I am a Senior Researcher at NatCen Social Research.

Adam Sharples: Hello, good morning. I am Adam Sharples. I used to be a Director General in the Department for Work and Pensions. I now have a number of non-executive roles, including Chairman of Ixion, which is a not-for-profit provider.

Chair: It is nice to see you here in a slightly different guise.

Adam Sharples: Thank you for inviting me.

Q2 Chair: Thanks very much for coming. Can I start with a very general question about how effective Jobcentre Plus is in its approach to face-to-face interviews with claimants as part of their employment support? Is five to 10 minutes the correct way forward? Is that the most efficient way of engaging with claimants, to get them into work? Who wants to start?

Tony Wilson: I am happy to start. As we have said in our evidence, we think that Jobcentre Plus, as a public employment service, has to play a number of different roles. It has to support the effective functioning of the labour market for people who have become unemployed, getting them back into work as quickly as possible. It has to maintain a rights and responsibilities regime, a conditionality regime, to make sure people comply with the requirements of being on benefit. It also needs to provide personalised and tailored support for people who have more significant barriers to work. The intervention regime, the five to 10 minutes a fortnight, attempts to tick different parts of that box. The evidence suggests that it is less than five to 10 minutes now, for the core fortnightly interventions; that was the original operating model for Jobcentre Plus. The DWP’s own research on the new flexibilities in Jobcentre Plus suggests that contact time is much less than that. In many cases, it is a box-ticking, tick-and-turn-type exercise. It is largely about enforcing that rights and responsibilities regime, ensuring that people are complying with the conditions of being on benefit, and that they are maintaining their job search and actively seeking work.

To that extent, it appears to be fulfilling its objectives. That first objective, around ensuring a flexible labour market and maintaining support, is harder to judge. However, it is hard to believe that a very short intervention every two weeks is sufficient to give people the support they need. Increased outsourcing of that to online job search and new channels presents particular challenges for many unemployed people.

Q3 Chair: Do you think it would be better if Jobcentre Plus had, or passed on to their advisers, a bit more flexibility about how they approach individual claimants?

Tony Wilson: A lot of that flexibility does now exist, since the reforms that were introduced in April 2011, which gave frontline advisers much more flexibility. However, while it led to more flexible use of personal adviser resource-that is the part around engaging people with particular disadvantages, which has been used much more flexibly-it appears that, on the whole, it has led to advisers spending less time with people on that regular, fortnightly face-to-face intervention. It has become more of a compliance-based intervention.

Fundamentally, there is an issue of resource here. There is a heck of a lot of evidence that Jobcentre Plus has been incredibly effective, and has more than paid for itself. The DWP evidence suggests that as well, with the introduction of Jobcentre Plus and the JSA regime. There is a return to the Exchequer, and a return to the economy as well, of engaging effectively with unemployed people and getting them back to work as quickly as possible. That resource has been reduced; there are no two ways about it. The NAO study published a few months ago shows that. The DWP’s own accounts show that. The research shows that. We think that reducing that resource is a false economy. There is fundamentally an issue of resource, and Jobcentre Plus, at the frontline, are having to make very difficult decisions about how they allocate an ever-shrinking resource, in terms of personal adviser time.

Q4 Chair: Do you not have some concerns, shared by the NAO, that it is very difficult to measure the success of increased adviser flexibility?

Tony Wilson: Yes, absolutely. There have been no attempts by DWP to measure the impact and the cost benefit of the reforms that were introduced in 2011. However, there is a wealth of evidence about what works in terms of personalised, face-to-face, quality adviser inventions. There has been a whole load of studies over the last decade or so, looking at Australia, the Netherlands, the UK and the US. It comes to conclusions around the size of caseloads, for example, and around the importance of personalised support. There is good research, including some by other people here, about how we can deal with it effectively. That is the piece that I am worried about. Yes, there is a particular issue about how Jobcentre Plus is measuring its own impact.

Q5 Chair: Could it be measured?

Tony Wilson: The impact of the flexibility?

Chair: Yes.

Tony Wilson: Absolutely. It could potentially be measured, if it was done in a controlled way. I would love to do this: if we went back to the regime that existed before 2011 in some parts of the country, which was a very structured regime, with personal adviser interventions of set durations, for different claimants at different points, and compared that with a more flexible regime, and tested different types of flexibility, you could absolutely trial the effectiveness and the added value of that. It is very hard now; it is much like trying to measure the impact of the Work Programme, with a black box operating model. It is very hard now, when we do not really know what advisers are doing, and DWP does not really seem to know what Jobcentre Plus is actually doing, and how it is using its flexibility, to measure impact in that way.

Chair: I have two colleagues who want to come in, and I will come back to the other witnesses.

Q6 Glenda Jackson: This is on the issue of the face-to-face interview. If I look at my own Jobcentre, for example, they are being very flexible with young people. They are meeting in groups and they are finding that beneficial. However, on the successful face-to-face, there are huge variations in claimants. Can you break that down, as to who are the most successful; who were, in one sense, the easiest or the most difficult? Do you know what I mean here? Some people are very, very difficult to get back into work.

Tony Wilson: Absolutely.

Glenda Jackson: Can you give us any divisions on who were the most successful under the old regime?

Tony Wilson: I should probably invite other members of the panel to speak. There is evidence about the relative importance of face-to-face advice for different groups. For example, there is work that NatCen has done, which has looked at well-being and mental health. From the research we have done with lone parents, we know about the importance of professional lone-parent advisers and good-quality support. I have flicked through some of the evidence, and there is a lot of it, submitted to this inquiry, and there were a lot of submissions from representative groups, from local authorities and others, about groups that have been particularly affected by changes in face-to-face support.

Clearly, it comes up again and again from research with young people, about the importance for them of quality time, particularly in the context of careers advice, and good-quality face-to-face careers advice. It is quite a mixed picture. It is something that we have not really tested systematically for a little while. A very good report was published in 2007 by DWP; it is an analysis of different research about what works for whom, which tries to get underneath some of that.

Matthew Oakley: There is a difference here between providing support and providing a monitoring and conditionality regime. For instance, there is no reason that we need to do the monitoring face-to-face necessarily. There is some evidence that shows that actually, it is just the act of going to a Jobcentre rather than seeing an adviser that acts as a spur to look for work, and to push people off benefit as the current measure does. It is that second element-the support-where you are completely right: some people need much more support and have far greater needs, and that is provided for much more effectively face-to-face.

Going back to your question, then you could have the flexibility to say, "Actually, perhaps we need to see fewer people face-to-face if they are very close to work. If they are just looking for jobs and they do not need much support in doing that, we can stop seeing them face-to-face, just bringing them in." We can then say, "That will free up a lot more time to give people the support they need face-to-face, in a much more personalised way, and with a longer timescale."

Q7 Glenda Jackson: Is it not the other way round at the moment, because of reduced circumstances in Jobcentres Plus? The more difficult ones are left to one side.

Matthew Oakley: Completely. Absolutely.

Glenda Jackson: Yes. Thank you.

Nilufer Rahim: Our evidence does support what Tony and Matthew have said. We evaluated the trailblazer for support for the very-long-term unemployed. This was two newly designed programmes by the Department for Work and Pensions, which tested the relative performance of the ongoing case management programme, which was more intense help and support from a Jobcentre Plus adviser to customers. It was delivered by Jobcentre Plus against a work experience scheme delivered by private contactors. On the ongoing case management programme (OCM), which we compared against the standard offer of Jobcentre Plus support, the OCM advisers did see customers much more often. Jobcentre Plus advisers can bring in customers as and when they wish, and the key to this was that they had much smaller caseloads. Their caseloads were typically around half the size of a standardoffer adviser. They also saw their customers for much longer durations. Just to give you an idea, 43% of jobseekers who were on OCM said that they saw their personal adviser weekly, and that compares to 6% who were on the standardoffer support.

There are a number of key advantages to that approach. It is mainly that seeing customers much more often helps to build up the rapport and the trust that enables customers to open up about their barriers, and any underlying issues, which perhaps earlier on they were slightly reluctant to bring up, such as alcohol or dependency issues. That greater insight into customers’ barriers then helps advisers to personalise support around customers’ needs more effectively. As Matthew said, advisers were able to go beyond the monitoring of job-seeking activities to providing a more active support, so helping people not only by sign-posting to employment or training opportunities, but helping with CVs and applications, and providing that one-to-one support that was really useful for customers.

Q8 Jane Ellison: My question is a fundamental one to Tony. You seem to imply in what you just said that the only way an organisation can be more effective is to increase the resource that goes into it. I just wondered if that is what you feel about Jobcentre Plus. I think you would agree that it is not universally true of every organisation.

Tony Wilson: Yes, I certainly would not want to give that impression, and I would not want to think that we should measure Jobcentre Plus’s performance based on the amount that we spend on it. My concern is that, fundamentally, for a business that relies so much on engaging people face to face and supporting them to overcome their barriers and move back to work, that does mean that you need to invest in the professionalisation, the training and the number of people who are delivering those sorts of roles. Caseloads for personal advisers have increased significantly. A survey that was done in 2009 by the University of Melbourne, which was looking across different countries but did look at the UK as well, found on average, there were personal adviser caseloads of about 80 per adviser. That figure is now 138 in the NAO report.

In a tougher labour market, there are much, much larger caseloads, and I think that does present real challenges for how effectively we are engaging people. We have seen big changes in how those resources are used. We have 15,000 advisers who are providing general advice to jobseekers, with specialist support as well. There are 1,500 providing support to lone parents, of whom there are 600,000 on Income Support, and only 500 specialist advisers supporting people on Employment and Support Allowance, of which there are 2.6 million. There are substantially more than the other two benefits put together.

I think there are issues about the level of adviser resource, and, of course, how effectively we use that. Regarding the efficiency challenges, Jobcentre Plus had done an incredible job in becoming a more efficient organisation, through better use of technology, better integration of back office services, much less duplication now between DWP and Jobcentre Plus services and much better use of resource locally as well. It has delivered incredibly well on the efficiency agenda over the last five years, which is demonstrated by the fact that it has not fallen over in the recession, frankly. However, I do think that there is a need for more investment in adviser resource.

Q9 Chair: Adam, you were in charge for some of that time. Was that down to the hard work you did?

Adam Sharples: I absolutely endorse what Tony says about the quality of what Jobcentre Plus does and what it has achieved, and a number of reports have been done that have looked back at its performance through the recession and through those challenges. Those reports have confirmed the quality of the response and the effectiveness of that response. The point I was going to make was that if you went out talked to people in Jobcentre Plus, as I am sure you do-

Jane Ellison: We just have.

Adam Sharples: I do not think you would find anyone who would not want to spend more time face to face supporting their clients. I think everybody in Jobcentre Plus would say, "If we could spend more time face-to-face, we could give valuable support and we could help people to move more quickly into work." However, the fundamental problem is that Jobcentre Plus, like any organisation, is resource-constrained, and it is really important to remember the volumes of people that Jobcentre Plus is having to deal with. Obviously, caseloads have gone up, but it is also very important to remember the flow through the system. There are 3 million to 3.5 million claims for Jobseeker’s Allowance every year, and perhaps half a million for Employment and Support Allowance. There are round about 4 million new claims coming through the system each year. The big challenge for Jobcentre Plus is to keep on top of all of that to fulfil the fundamental requirements of the system, that there should be some check that people meet the conditions for receiving benefit, and then also provide some support for people.

I have to say that one thing that really impressed me when I saw Jobcentre Plus working was the quality of the staff who were working with customers and the quality of the organisation in Jobcentres. I always found it was a very calm and well-ordered atmosphere, even though there were sometimes difficult people coming in and quite challenging conversations to be had. It was all handled very, very smoothly. So it is that dilemma: how do you get face-to-face time, which everybody wants, within constrained resources?

Q10 Chair: One of the recommendations we made in our inquiry report on the Work Programme was that perhaps Jobcentre Plus needed to introduce the Australian classification system. They have that instrument so they can identify those who have the highest barriers or the most barriers to work. Do you think that is something that we should be considering in this country?

Matthew Oakley: Certainly, moving more towards that approach, where we are introducing a classification system that says, "How far away from the labour market is this individual person?" and then tailoring the support and conditionality requirements based on that assessment is a key thing we should be doing in this country. I would say that, actually, we need to go further than what the Australians have done. Tony probably knows the evidence better. There are arguments over whether their classification tool is as accurate as you might hope a tool might be. With modern-day data technology and what we can do now, across the private sector as well as the public sector, in terms of splitting caseloads of customers of different companies, we should employ the same kind of techniques in Jobcentre Plus. We should bring together data from other Government Departments, along with a questionnaire of the individual, bring that together with private sector data, which is used in other public services, to say, "What’s the likelihood of this person reaching long-term unemployment?" If that is very high, then on day one of the claim they should be getting the support that they need from providers like the Work Programme, rather than having to wait for 12 months in Jobcentre Plus, where we have already seen that, frankly, they can be parked without the support they need, because it is just a big sausage factory. The conditionality is fine, but you are not getting the kind of personalised support that you need, so we should be putting in place a tool that does that on day one and says, "Which of these people need support right now?" and then send them across.

Q11 Chair: Are there any cons to that process?

Matthew Oakley: Of course there are. The criticism would always be-and you would hear this from people at the Treasury-that there is dead weight associated with it. You get false positives; you think someone needs help, but actually they do not, so you give them support on day one that they do not need. We are currently doing that in the existing system. We have segmentation in the existing system that is based on benefittype, essentially, so people who are moving out of prison go on day one, and other people are fast tracked at various points throughout the system. We never measure what the false positives of that system are, so we do not have an effective baseline against which we can really measure the success of a new potential tool. So I push the Department to work out what they are currently doing and how effective that is at identifying those people with the greatest needs, and then you can compare that to a new system and see whether it is more effective or not.

Q12 Nigel Mills: That all sounds quite attractive. Has anywhere else in the world got that system up and working, or is this completely blue sky?

Matthew Oakley: A number of other countries have systems similar to Australia. In the USA, some states do it, and the Netherlands do it as well. Germany do some of the stuff, too. If we move to this new level, that would be pushing the boundaries of where countries have gone before. However, I do not think there is any reason not to try that. Again, you could do this on a pilot basis in one Jobcentre district. If it works, fantastic. If it does not, then you have not lost anything.

Q13 Nigel Mills: When do you think that initial contact needs to be? I think you said when the claim is first made, but that is not necessarily when somebody first loses employment, is it? If I have a redundancy pay-off, I may not go to the Jobcentre for a couple of months, thinking I will find work under my own steam. You can lose two months of the most enthusiastic job-searching before we even know them. How do you get that intervention starting when it needs to?

Matthew Oakley: I guess that is difficult, because obviously you have little control unless someone comes forward and says, "I’d like benefits," and then, "Okay, you can have benefits, but we need to ask you some questions and think about what kind of support you’d need." Perhaps part of that is trying to remove some of the stigma around claiming benefits, so people do come forward more quickly to claim benefit when they become unemployed.

Stephen Lloyd: I am sure that will delight the Treasury.

Q14 Glenda Jackson: I quite agree with Mr Sharples on the way Jobcentre Plus has stepped up to the plate, but what is the basic motivation here? Is it to actually get people work, or get them off benefits? They are not the same thing. Ministers have said to us that the majority of claimants are back in work within six months; it is a very easy churn. They told us that the whole process is to concentrate on those who are hardest to reach, and yet we all know that that is not coming. Is the underlying motivation not necessarily to get people into work, but simply to get them off benefits?

Adam Sharples: Could I say a word on this? I think there is quite an interesting history here. It seems to me the motivation is absolutely to help people into work. The measurement that is used is a flow off benefits, but the reason for that is purely pragmatic. That is the thing that the Government can measure, when somebody leaves benefit.

If you wind the clock back six or seven years, Jobcentre Plus used to measure entries into work. It had a team of several hundred people who would phone up employers, and find out whether somebody was in a job. Jobcentre Plus had a "job-entry target". The general view was that that was pretty inefficient, having all these people making these phone calls just to check up. So Jobcentre Plus moved to what was called a "job outcome target", which tried to use the links with HMRC to pick up on PAYE data and cross-check against benefits data, using national insurance numbers, and use that information to find out when somebody was in a job. The problem with that was that, because of the nature of the PAYE system, there was a long time delay, so the data was always out of date, and you could never track it back to an individual office. Jobcentre Plus felt that it was a very, very poor tool for actually managing and motivating staff, because they did not know in that office how many people had got into a job.

It was at that point, just for pragmatic reasons, that Jobcentre Plus switched over to measuring what they could count, which was people leaving benefits. It is a second best-everybody recognises it is a second best-but it is a good enough measurement of what that Jobcentre is achieving. The challenge is to try to constantly remind people that what Jobcentre Plus is there for is to help people get a job, not to get people off benefit, but the measurement is a pragmatic proxy for numbers of people getting into a job.

Q15 Glenda Jackson: But removing people from benefit is no longer a choice for Jobcentre Plus, is it? That is something that the present policies have intrinsically within them.

Adam Sharples: It is certainly true that Jobcentre Plus is targeted for the off-flow from benefit after six months: what proportion have left benefit at the six month point. As I say, though, that is the nearest existing measurement that any Government agency can get to what it would really like to get to, which is: "How many people have got a job?" Other countries have different systems. The US, for example, has a job hire database, and they can immediately measure when someone is in a job. We have not, but perhaps with the real-time information (RTI) system that is providing the foundation for the Universal Credit, there will be new opportunities to measure job entries more accurately. I would hope that the Department is looking closely at how it can move to a job-entry target or performance measurement as quickly as possible, rather than an off-flow from benefit.

Q16 Chair: I understand, Matthew and Tony, that you have quite strong views about the measurement that should be used. Do you agree with that, or do you think there should be a different measure?

Matthew Oakley: I completely understand where Adam is coming from in terms of the pragmatic approach and measuring things that are easy to measure and which, in terms of resources, can be done quite cheaply. That is in the past. Now, with RTI coming in and the link between HMRC data and DWP data coming into place, there is no reason why we cannot change that, particularly as we are already asking Work Programme providers to do exactly what Adam is talking about in terms of matching people going into jobs and then keeping track of them when they are in jobs. That is the point: we are already trying to make Work Programme providers do that, so why not Jobcentre Plus as well? As I say, the link between HMRC and DWP data is going to be hugely beneficial to that. When the Department has done this in recent studies, it has shown how ineffective Jobcentre can be in helping people stay in work. So, if you look at a recent DWP publication and use that with the offflow data, it basically shows that only 36% of people move into Jobcentre Plus and then find a job within six months that lasts for eight months, so a sustainable job within six months; only 36%. Compare that to the 50, 75% offflow and it is a very different picture indeed. That is the kind of performance measure we are expecting from the Work Programme and then say that is failing compared to Jobcentre Plus. So it does not really seem to be a particularly fair assessment at the moment, which is why we need to move to a system where we are checking whether people are moving into work, how long they are staying in work for and how much they are earning.

Tony Wilson: My views probably are not quite as strong as Matt’s on this, because the job outcome target, essentially, as the Department said at the time and also when they see you, I expect, was like driving a car by looking in the rear-view mirror, because they could only judge performance with a oneyear delay. It simply did not work and now the focus must be about how we can use technology, realtime information to have a job-entry target. Work Programme providers have a job-outcome target. They also have to invest a heck of a lot of resource, phoning up employers and asking if somebody is in work and making all those checks. It is completely disproportionate to what they should be doing. The Department had originally intended there would be an IT solution, which would mean they would not have to do this, but that has not worked, so it is very labour intensive.

The DWP research on this, as Matt knows and has said, shows that about twothirds of people who leave JSA do leave for work. I would not necessarily agree that that is effective or ineffective. What we do not know, because there are no comparable studies, is whether that is good or not. What we do know is that in 2004, when there was the same kind of survey, it was only 63% or 64% of those leaving benefit who were finding work. The likelihood of finding work has got slightly better over the last decade if you compare the 2004 and the 2011 survey, so it is hard to know whether that represents good or bad performance. The other thing it shows is that fully 80% of people who leave JSA and find work are still in work seven or eight months later. Now, you could present that, as Matt has done, as a bad figure, but equally, if you present it to a Work Programme provider, 80% of people who enter jobs will still be there in six months, they would bite your arm off. They are nowhere close to that sort of conversion rate and nor would they be, because they deal with a harder-to-help group, frankly. But it is quite hard to make these comparisons, because we just do not know enough about what good should look like. I think Matt is absolutely right; we do need to test different approaches. We need to look at different ways of measuring it, but we also need to think about how we benchmark what good looks like for getting people into work off JSA.

Nilufer Rahim: I would add to that by saying that our challenges that are being experienced in recording destinations of offflow can hide some potentially unwanted or negative results. For example, when we did an analysis of offflow in our LongTerm Unemployed Trailblazer evaluation, a small minority of people left benefits completely and did not enter work. When we looked at that proportion of people, we found that after those who went on to Employment and Support Allowance, they had the highest incidence of mental health issues. They were also single, so not living with a partner. They were also most likely to have had their benefits stopped in the past six months. When we had qualitative interviews with people who had come off benefits but did not have the financial contingencies of a social support network and of paid work, we found that there were severe financial repercussions for these people, which is unsurprising. But this nonrecording of offflow destinations does muddy the picture. We do not know whether offflows are a good thing or a bad thing.

Q17 Mike Freer: We have covered some of the issues, but I just wanted to go back on to the monitoring of activity. We visited some Jobcentres yesterday and the day before. We saw the monitoring of the Claimant Commitment and had they done their 30, 32, 34 hours of activities. If you argue that what gets measured gets done, that is a good process. On the other hand, you could argue, if it is not effective, what are we learning from it? So should we be continuing with the "how much have you done?" or should we be moving to a "how effective is what you have done and what can we do differently moving forward?" approach? Which do you think would be most effective?

Adam Sharples: My view very strongly is that the organisation has to be focused on what results it is achieving, and has to have a performance system that gets each member of staff focused on not "have I completed this bit of process?", but "has that process been successful?" If you look back over the last five years or so, there has been a real shift in Jobcentre Plus. Five years ago, it was very focused on making sure everybody did the right bits of process. I can recall a lot of conversations I had in Jobcentre Plus where people would tell me a lot about the steps they had to go through in a discussion with a customer, but when I asked, "What difference is it making for that customer?" there was a slightly blank look.

What we did when we changed the performance regime was to really concentrate it on-I know it is not a perfect measure, because it is offflow from benefits rather than entries into jobs-the result. We tried to give much more flexibility to Jobcentre Plus staff to think, in any particular interaction or conversation, "How is it going to help this person to get into a job and therefore get off benefits?" My sense is that that has been a tangible shift in Jobcentre Plus thinking and a very welcome one, but it can still be pushed further.

Tony Wilson: I absolutely agree with Adam. There is an issue in that fortnightly compliancetype intervention, that fortnightly face-to-face job review, about how far advisers are able to do that in what is a couple of minutes intervention now. This is the challenge. So with the Claimant Commitment under Universal Credit the expectation will be that job seekers are undertaking activity that is equivalent to fulltime work, essentially, 40 hours or 35 hours of activity a week. Immediately, the discussions within the Department and with outsiders like us were around, "How do we operationalise that? How can we make that bit of process, 35 hours a week, into something that we can give to advisers that they can monitor on a fortnightly basis?"

If they are spending two or three minutes with somebody, it is so hard to get that focus on what is the impact and the outcomes that you are achieving. But there are pointers on this: Jobcentre Plus piloted in Loughton in Essex, with the Behavioural Insights Team from the Cabinet Office, a different form of intervention with more time, a different form of Claimant Commitment: a claimant agreement that was more "set out what you are going to do, verbalise what you are going to do, come back two weeks later and discuss whether you have done it or, change your objectives, go out and do it again." That sort of engagement, which is much more focused on one-to-one engagement and then reviewing what people have done, how effective it has been and then changing the agreement and going out and looking for work again, did get better results. That goes back to the heart of where the JSA regime started in the late 1990s. Where Jobcentre Plus started in 2001. I sound like a broken record, but that is the point we do risk losing and we have risked losing in the last couple of years with the very high volumes and really a focus on that fortnightly intervention of it just being compliance-based, because then it cannot get away from box ticking. It cannot get away from just saying, "Have you done 10 activities? No, you have not, so I will sanction you for not actively seeking work."

Q18 Mike Freer: There are two follow ups on that. Firstly, "nudge" of course is the other term for behavioural insights. Does that allow the adviser to either focus on the barriers-so lone parents or disability-but also then to flex whether it should be a fortnightly intervention? If I turned up at the Jobcentre, they might say, "Mike, you are motivated, you know what you are doing, I will see you every six weeks." Other people might need a weekly support session, so are you suggesting that flexing that approach would allow it to be more tailored and effective?

Tony Wilson: Absolutely. There is potential to do that and that is exactly what we should be testing. Indeed, the DWP is doing some tests on that, which I think concluded in March and, hopefully, we will start to see some results from. But there will always need to be a regular engagement that is a check to make sure that those individuals are still meeting the conditions of being on benefit and that their circumstances have not changed. That is an important statutory responsibility of the Department. But yes, that greater flexibility is important. One caution on that, though, is that there are about 4.5 million claimants of outofwork benefits and the very large majority-twothirds-are seen sixmonthly if at all by Jobcentre Plus.

People on ESA, which make up the large majority of people on benefit, and people on Income Support, do not really have any engagement with a personal adviser. We have not discussed it and it has not really come up a huge amount in the evidence to the inquiry, including the written evidence. That is the group that is at the biggest risk of being parked: 95% of people on ESA are not in the Work Programme, are not in Work Choice and if they are seen by Jobcentre Plus they are seen every six months. That is a group, among others, where we should be increasing resources, looking at how we can use resource more flexibly. If that means, therefore, spending less time with more employable job seekers, those are the trade-offs we need to consider.

Q19 Sheila Gilmore: We have just been on a visit to AshtonunderLyne to see what they were doing with the Universal Credit first applicants. They did make a lot of a system of going through what people would do in the next fortnight, but it did seem to me that that was quite timeintensive. Is that feasible when what a lot of us are hearing and have been hearing for the last couple of years at least, if not longer, is people getting four minutes, five minutes, in, out.

Adam Sharples: As I understand it, Jobcentre Plus does quite a timeintensive discussion when somebody first signs on, typically 45 minutes to an hour, and that has two purposes. One is to try to establish a Jobseeker’s Agreement: what that person is going to do to look for work. The second is to check that all the details of the claim are right, so that the money can be put into payment. That is quite an intensive process upfront. It then goes into a phase where the customer is seeing Jobcentre Plus every fortnight and for a very short period. Jobcentre Plus is trying out lots of different approaches to that fortnightly signing, including electronic signing, speed signing and all sorts of other techniques. Interestingly, it has been trying to do that using a random assignment, so it can genuinely see what makes a difference in a particular office in a particular labour market. It will be fascinating to see the results of those trials when they come through; I think they are due later this summer. That is one thing that they are trying out.

Linking back to something I was saying earlier, one thing where there has been a shift in Jobcentre Plus is, previously, the person you saw when you came in for your fortnightly signing was just random. You could see anyone, so there was no relationship built up. Increasingly, Jobcentre Plus has tried to get a personal adviser working with an individual. If you are seeing the same person each time you come in, you are much more likely to feel a sense of obligation, to feel that you need to tell the truth; there is a relationship there. That is hugely welcome. I hope that what Jobcentre Plus can do looking forward is to think about how that slightly more personalised approach can be pushed further. I know it is going to be horrendously difficult within the resources available, especially given today’s expected announcements about spending constraints, but the more personalised it can be the better.

Q20 Stephen Lloyd: JCP offers a wide range of employment support under the brand name "Get Britain Working". It is across a range of particular areas, and there are three I would like to focus on. One is the work experience. Another is the New Enterprise Allowance, which is a cunning reuse of the name "Enterprise Allowance" from the 1980s, but I think that is very effective. The third is the sectorbased work academies. Looking at Nilufer, if that is all right, because I know you have done some research in this area, what is the data showing in each of those three areas initially? Is it looking to be effective, not effective, middling effective, very effective, in work experience, New Enterprise Allowance and sectorbased work academies?

Nilufer Rahim: Our analysis did not look at those three specifically and then look at job outcomes or impacts associated with those programmes. Those measures were there for Jobcentre Plus advisers who were delivering the ongoing case management programme-that more intense form of support-to draw on. What they discussed was their ability to use those measures more effectively, and the key to this was seeing people more, understanding their barriers more and so timing those forms of support around people’s needs more effectively. What we did find was that, on OCM, those who received that more personalised support, so seeing their adviser more often and doing the types of activity or those sorts of measures that were most relevant and tailored to them, were more likely to get a job outcome.

Q21 Stephen Lloyd: What about the outcomes when people did the work experience, because it is almost 100,000 people now? Did you do any research on that, or Tony’s side?

Tony Wilson: DWP has published an impact assessment on work experience, which found that it had a positive impact on offflows from benefit and, as I recall, on entries into employment as well. That was using essentially quasiexperimental methods to control for other things that could influence the likelihood of somebody leaving benefit. That was quite surprising and welcome, because the work experience programme that DWP delivers is incredibly cost-effective. It is very, very low cost. Essentially, there is a work experience coordinator and people are referred into a placement where there is no payment to the individual or the employer. It is essentially done as a public good by employers and it has had a positive impact. If you ran a cost benefit analysis, although we have not, I think you would find really quite positive impacts from it.

That goes to the heart of one of the critical barriers that young people face, which is around the signal that being young and unemployed sends to employers: that you are not employable, that you are not worth giving a chance to and the fact that young people often end up near the bottom of the very large pile of job applications. Work experience helps to overcome those signalling barriers. It builds people’s confidence, gives them access to the networks we all use to find new jobs and new opportunities, and it improves their workplace skills. It compensates a lot for the fact that work experience in schools is often low quality and is no longer a statutory requirement and that, increasingly, young people in education do not work. The employment rate of young people in education has halved. I think I have said this before in a previous inquiry on young people, but the employment rate of young people in education has halved over the last few years and the employment rate of young people full stop, 16 to 17yearolds, has halved as well.

Q22 Sheila Gilmore: The pilot that is always referred to in terms of figures is it still that 1,300 that was done nearly two years ago? Has there been any further proper research on work experience as opposed to anecdotal?

Tony Wilson: The DWP impact assessment was published about a year ago. I am not sure how many people were in the sample, but I think you would find similar results. I have certainly not picked up any indications that the offer or its application has changed. But there is an important point there more generally about Get Britain Working, which is that there is no evaluation plan for the Get Britain Working measures. DWP is not evaluating them in and of themselves. The work experience measures should come up in the Youth Contract evaluation that DWP will publish this year. The New Enterprise Allowance is being evaluated, and I think that will also be published this year.

Q23 Stephen Lloyd: What about sectorbased academies?

Tony Wilson: There are no plans to evaluate sectorbased work academies. If you look at the evidence that the Association of Colleges submitted, they talked about how well this was going and how their partnership working with Jobcentre Plus had improved and how sectorbased work academies were a success. I have a soft spot for them, because I was in DWP and had some responsibility for them when they were set up along with the work experience programme. I just wish we would evaluate them. There is potentially a really interesting and important model here about how we can get localised college funding and local flexibility for Jobcentres working together to join up preemployment training, work experience, guaranteed interviews and, potentially, jobs. That needs to be a critical part.

Q24 Graham Evans: I just want to mention a success story, which I have mentioned before. I did a jobs fair at Mid Cheshire College with Jobcentre Plus. You had 17 and 18yearolds, a big cohort speaking directly to 40 private sector employers with good-quality jobs and apprenticeships, plus all the rest of Jobcentre Plus clients. It was a fantastic occasion and Jobcentre Plus and the College said, "Can we not do this more often?" The mix was just right: private sector good-quality jobs, which is the key ingredient, but the quality of Jobcentre Plus management and leadership along with the college leadership was a magical formula. We could and should perhaps put that into every community.

Tony Wilson: There is some not so good practice, but there is some really good practice coming out of a lot of colleges and the further education (FE) world now around how they are using their increased responsibilities locally.

Q25 Stephen Lloyd: There are and I can add to that. I am Chair of the APPG on FE and Lifelong Learning and we are doing an evidence session at the minute with colleges, employers and Jobcentre Plus, so I pick up on that.

I will come to Adam for a minute, simply with your hat on from a few years ago. You remember there was an enormous kerfuffle around work experience in the media two years or so ago. For a lot of employers and people like me who, before I came into politics, employed a lot of people, work experience always made sense for the reasons that Tony has mentioned. But I do take my colleague Sheila’s point around what counts as evidence. Would you agree with Tony’s illustration that the DWP really does need to put in a proper process of robust measurement around, say, those three that I am particularly focusing on-work experience, New Enterprise Allowance and sectorbased academies-and that without having relatively up-to-date data every six months, so to speak, it would undermine any potential success they would have? In other words, how important do you think the evaluation is on a regular basis?

Adam Sharples: I would strongly agree with everything that Tony has said. I would strongly agree with you that evaluation is important. Some simple data would help and it is a little bit of a puzzle to me, for example, on the New Enterprise Allowance. It seems to me this is a great success story. It is a fantastic programme that takes people who are unemployed, helps them set up in business, gives them a mentor, usually some training and also access to a loan to get started.

My business, Ixion, is delivering the New Enterprise Allowance in three areas of the country and we are so excited about the results we are getting for people. For every 100 people who come on the programme, 60 are getting a decent business plan produced and 55 are starting up in business. That means that just in our three areas of the country there are over 3,000 new businesses operating that are being run by people who were previously dependent on welfare. That seems to me a fantastic achievement.

Q26 Glenda Jackson: Are they just 16 to 18yearolds?

Adam Sharples: No, they are all ages.

Q27 Chair: Can I just clarify from what you are saying, though, there is data on work experience, but there is no data on any of the other measures under Get Britain Working? As a whole, do we have the data to tell us whether they are effective?

Adam Sharples: We do not.

Q28 Chair: Whose responsibility should it be to get that data? Is it Jobcentre Plus?

Adam Sharples: On New Enterprise Allowance, since providers are paid on the basis of what they deliver-if you get someone to the business plan stage you get a certain payment; for a start up in business you get a certain payment-someone somewhere in the system must have some data, because they are making some payments.

Tony Wilson: There is a statistical summary for Get Britain Working. It certainly publishes the work experience figures and, I believe, the sectorbased work academy figures. It may well publish the New Enterprise Allowance ones as well. But it is literally the headline, how many people have started in that reference month. What we are missing is the richness of data about what is being delivered where, how many people are benefiting and in what ways.

Q29 Stephen Lloyd: That is something that we as a Committee, if we agree, should be recommending very strongly, that the DWP comes up with that data and the detailed data.

Matthew Oakley: This is a strange position for me to be in, given that I am an economist and I love data analysis, but I do not think it should be DWP’s sole responsibility to be doing this. DWP is currently the largest employer of economists in the country, or something, and to make them evaluate every programme they have on their books every six months would increase that massively.

Q30 Stephen Lloyd: How do we square that circle then?

Matthew Oakley: My point of view is that there should be processes put in place to publish the data and the offflow data publicly, so I and Tony can evaluate it and get a bunch of academics to evaluate it without paying them. That has to be the way to do this.

Q31 Graham Evans: Good idea.

Matthew Oakley: This has to be the way of doing it. You publish the raw data with confidentiality stuff sorted out.

Tony Wilson: Someone has to pay us, Matt.

Chair: But you agree that there is a hole; that data is not being collected and it is not being analysed at the moment and it needs to be.

Stephen Lloyd: Or possibly it is being collected, but it is not being put out there. I think that is a really good point, so maybe it is us pushing them to be more open about the raw data and then leave it to you lot. Okay, moving on, Nilufer, one of the issues with Get Britain Working, which again I know you have done some research on, is the very longterm unemployed, the cohort that we have already mentioned.

Chair: Can I ask you to move on to question eight, because we have probably covered most of that?

Q32 Stephen Lloyd: Okay, that is fine. The Flexible Support Fund again, like most of the MPs around the table, I have had constituents in. A classic example is an HGV licence: "I want to be an HGV driver. I do not have the money to do the training. I have heard of the Flexible Support Fund, yet I go to my Jobcentre Plus and they are not interested in giving me the money." Clearly, there need to be parameters around the Flexible Support Fund. You cannot just hand it out willynilly, but if I can go to whoever thinks they are the right one to answer this, the JCP districts have access to FSF. How well do you think the JCP is doing in managing an effective use of the Flexible Support Fund and evaluating its impact? Who wants to go?

Tony Wilson: The answer is not at all well. The NAO report and the PAC went into this as well, and DWP acknowledged in that process that the only information that is collected is expenditure on the Flexible Support Fund. We do not collect data on how many people then benefit from particular support within the fund-what outcomes they achieve-and that would be just the very, very basic data, let alone getting in to trying to understand the impact of it.

Q33 Stephen Lloyd: And yet the JCP would have that data.

Tony Wilson: They may have that data, but they are not required to report it to the Department. So they may not have that data. This concerns me hugely. We do not even know how big the Flexible Support Fund is. There is no public record of the size of the fund. There was a note by the House of Commons Library, which estimated it might be around £200 million a couple of years ago, but DWP has not even said how big the fund is. We do not know what the allocation is in different districts. We do not know what the criteria are for applying for Flexible Support Fund grants. We know that it cannot duplicate provision and from the evidence you have received there are really mixed views about that. I think it was Hackney said that they put in a bid that did not duplicate provision and were told it could not be approved because it was going to duplicate potential future provision that Jobcentre Plus might provide.

There is no transparency about how much money is being spent. There is no transparency really about how to apply for it. There is good practice and bad practice in that. There is no transparency about how it is used, let alone about the impact. In terms of oversight of public expenditure, there is a real worry and concern here. There is a substantial chunk of Jobcentre Plus money, and we do not really know what it is being spent on. The DWP evidence talks about it being spent on travel, childcare, which is important, and other support for disadvantaged groups. Let us open it up.

Q34 Stephen Lloyd: On the note that Tony is making, do any of you have anything to add to that or do you broadly agree with Tony that there is a lack of transparency on it, a lack of clarity around boundaries? Is there anything you want to add?

Adam Sharples: I strongly agree with Tony. The one point I have a little bit of sympathy with the Department on is that we are always asking for more data to be collected and the Department may feel their job is to help their customers, not just have legions of people collecting data. I have a bit of sympathy with them on that, but I do think that the whole system, not just this particular element of the system, is remarkably opaque. Perhaps I feel this rather acutely having been inside the system for some time, now being outside and trying to read what is going on. It is very, very difficult to get systematic, clear information about how money is spent in the Department. A plea from me to you, the Committee, would be to be much more demanding, to ask for clear information. I do not know whether others have tried to pore over the Department’s accounts and work out where the money goes.

Q35 Stephen Lloyd: Is this a historical thing along the lines of Departments generally do not really want to have the light of raw data shined on them too much, is it something that is just too much hassle, or is it "No, let us keep it a secret"-in simple terms, cock up and conspiracy? What is your instinct? The word "conspiracy" is absurd in this situation, but you know where I am coming from.

Adam Sharples: I do not want to speculate on what the reasons for it are, but-

Q36 Stephen Lloyd: It is historical. This is not new.

Adam Sharples: I would argue that, paradoxically given the stated Government commitment to greater transparency, the trend has been for less useful information to be put in the public domain. For example, up until three years ago, every Department would publish a departmental report each year. That had a series of tables that had to follow a set format. I know a little bit about this, because when I was in the Treasury we defined that format quite carefully, to try to make sure that each Department published tables showing where the money went in a way that corresponded to recognisable administrative boundaries. So you could say what Jobcentre Plus was spending, for example, or how much was being spent on the Flexible Support Fund. Now no such reports are published. Jobcentre Plus itself used to publish an annual report. Now Jobcentre Plus does not exist as an agency and therefore there is no report. So, in a number of respects we have moved backwards. There have been some positive steps in other areas. My plea is just go on pressing, because in there, there is lots of data. Everybody has budgets in Government. That is the way Government works. I see no reason at all why those budgets should not be available publicly.

Q37 Nigel Mills: I think one programme Mr Lloyd did not mention was Work Choice, which obviously helps a relatively small number of people who need some intensive support. Do any of you have any views on how effective Work Choice is, whether it should be scaled up? Mr Wilson, I think you are doing some work on this, aren’t you?

Tony Wilson: Yes. My organisation is evaluating Work Choice and that evaluation will be published imminently. I am not personally involved in it and have not seen any of the findings, so it would probably be inappropriate for me to comment on it. Looking more broadly at the evidence surrounding disability employment support and services, there are very strong arguments in favour of specialist support for disabled people. There is some good OECD work about this. There are very clearly defined barriers and issues that many disabled people have and that would support the case for a specialist programme.

My one concern is, if you look purely at the numbers of people who are referred on to Work Choice, it is a relatively small programme of tens of thousands rather than the million-plus who are on the Work Programme. Many of them are not on Employment and Support Allowance; most of them are on JSA and many of them are on no benefit at all. There are 20,000 people who have been referred on to Work Choice who are on Employment and Support Allowance. So Work Choice still does not address that issue about there being 2.6 million people on Employment and Support Allowance; 400,000 of those are in the support group. There are 2.2 million people on ESA or IB who should be capable of some workplace activity and fewer than 150,000 of those are in any kind of structured employment support. But I cannot really comment on Work Choice specifically.

Q38 Chair: Do any of you know how JCP advisers decide whether to refer someone to Work Choice or into the mainstream Work Programme?

Tony Wilson: The evaluation of the Work Programme published by a consortium, led by the Institute of Employment Studies-and we are part of that consortium as well-looked at this and found some real issues around that referral process and those handovers.

Q39 Chair: It is a bit ad hoc.

Tony Wilson: Exactly. It also found a real lack of clarity about in what circumstances people should be referred to the Work Programme, versus Work Choice. The design of both programmes in some ways has contributed to that. For example, the random allocation nature of the Work Programme means it is quite hard to actively refer a volunteer into the Work Programme because you do not know which provider they will go to when you are trying to build relationships with them. The fact that it is a twoyear programme means that it is a heck of a commitment to make. All that may have tended towards people being more likely to be referred into Work Choice, but it is quite opaque. As I say, the fact that the large majority of people who are referred into Work Choice by Jobcentre Plus are on JSA I think reflects that those are the people they see. There are 2 million people who may be capable of workrelated activity on ESA who, frankly, are not seen by advisers and do not get the opportunity, unless they volunteer themselves, to go into Work Choice.

Chair: That is something we are going to have to explore a bit more.

Q40 Sheila Gilmore: Is that because the workrelated activity group (WRAG) section of ESA claimants are not getting much time from JCP? Constituents of mine report that they go into the WRAG, they are called in for one workfocused interview and told to come back next year.

Tony Wilson: Come back in six months, exactly.

Sheila Gilmore: Six months or sometimes a year.

Tony Wilson: It is twice a year. I just feel like we have really gone backwards here. Since 2005, successive Governments have tested and tried to test and develop new ways of engaging with this group-through condition management programmes, through supported employment, through wage subsidies, through more effective engagement with employers, through different forms of adviser support, different professional forms of advisers-through some really innovative stuff that has been tested and proven to work in randomised controlled trials in the US for individual placement support services and supported employment-type provision as well. We had that whole process of testing through pathways to work and everything else, and we have largely stopped it. We have chucked it out, largely. That is partly because of the recession and partly because of the need to reduce expenditure on programmes. But when we come out the other side of the downturn, we are going to find that we have a large number of people, the very large majority of people who are on benefit, who are a long way from work, are on ESA and have not received any meaningful support for a substantial period of time.

Matthew Oakley: It is also a wider reflection of what you were saying around referrals to the Work Programme and other schemes more generally. If you look at Payment Group 3 on the Work Programme, which is early referral rates from JCP, by Jobcentre Plus district in our recent report, as a proportion of the claimant count that varies from less than 5% to above 20% in different Jobcentre districts. There is no clear reason why that would vary quite so much. So, as Adam was saying, the whole system is quite opaque at the moment about how it works; it is not particularly clear.

Chair: We will now move on to Universal Jobmatch.

Q41 Anne Marie Morris: If I may, Adam, I am going to focus on what it is rather than how it is used. I have a quick question for you. There is some concern that there is no checking of the vacancies on it. Some of them are a bit iffy, a bit dubious. Is there a process in place to ensure that the jobs there are genuine jobs and are what they say on the can?

Adam Sharples: I cannot really comment on how exactly the vacancies are fed into the system. This was put in place by Jobcentre Plus after I left the Department. But having tried it myself, and I am now registered on Universal Jobmatch, it seems to me to have the basis of a very useful and quite sophisticated system. It allows you to register, put in what sort of jobs you are looking for, what sort of area and it will come up with useful vacancies that match your criteria. It has the capacity to link in to Jobcentre Plus advisers, so that if you want it to tell Jobcentre Plus when you have applied for a vacancy, it can do that. You can store your CV and make job applications very easily. It is a great system. I would argue it could be made more sophisticated, because the information it holds about you, your search criteria, is pretty crude and if the sort of work you are looking for does not fit the categories, then it does not quite work for you. I can think of all sorts of ways it could be made more useful, but it is a great start.

Q42 Anne Marie Morris: Okay, that is helpful. Tony, can I ask you about the issue of the lower-skilled vacancies? There seems to be a concern that those are not there, whether that is because that simply is not the way those sorts of jobs are advertised. What can we do to ensure those low-skilled vacancies are available?

Tony Wilson: We talked about this in our evidence. Firstly, I would echo Adam’s point. We think Universal Jobmatch has the potential to be a huge improvement on what has gone before and to really transform the way that unemployed people are supported to find work. Essentially, we have had a system in the past where Jobcentre Plus collected a fraction of the vacancies in the economy and that fraction tended to be congregated around lower skilled, entrylevel jobs, and even then it was not picking up all of them. Universal Jobmatch is intended to keep that, to bottle that and then also to get potentially middle and higher-end vacancies as well, managerial vacancies. It has certainly succeeded in the latter.

What we are concerned about is that, in the analysis we have done looking at the profile of job starts against the profile of Universal Jobmatch vacancies and the profile of previous Jobcentre Plus ones-and we had to get this from EU data because it is not published by the Department; it was part of an EU report-it does look like there has been a really significant dropoff in the number of vacancies that are being captured in elementary and lower-skilled jobs.

Q43 Anne Marie Morris: Any idea why?

Tony Wilson: It is not clear. It may be about how it is being recorded, because one issue with Universal Jobmatch is that the data is not being recorded in any industrial or occupational classification that most people would recognise. It is very hard to match the vacancies that are in Universal Jobmatch to the standard industrial and occupational categorisations, so it could be partly about that. It could be that something has been lost in that transition from the Jobcentre Plusbased system to Universal Jobmatch. You would have to ask the Department, but it does look problematic.

Like Adam, I am registered on Universal Jobmatch. I went on and put in my profile as if I were someone looking for sales and retail jobs in London. I was offered a job in Bracknell as a retail solutions consultant; that was in the top 10. The other top 10 jobs included a sous chef and a series of managerial roles in retail and something called a "copy and print centre specialist". There are so many retail and sales jobs and it is partly because you cannot segment by your own level of qualifications or skills; you cannot segment by your previous experience or particular employers, for example. There are so many jobs, but it is generating this managerial stuff.

The fundamental issue for us is that it is being used more and more as a means of enforcing conditionality. People are required to register on it. If they do not register on it they can face sanction. Jobcentre Plus can monitor activity on it. They can require you to apply for jobs through it. They can follow up what you have and have not done through it. We would argue that that is not how jobseekers are going to find vacancies right now. You really have to consider, if Universal Jobmatch becomes the main or an important way in how conditionality is being applied right now, that is a really serious concern. It needs to be a tool to help people find work. It cannot be a means of checking compliance and ultimately sanctioning people, because if they are finding jobs through their mates, their neighbours and friends, then whether or not they are registered on Universal Jobmatch is neither here nor there.

Q44 Anne Marie Morris: That is helpful. Matthew, taking that point to the next stage, do you think there is a risk that there could be too much dependence on using technology to find jobs and that people are going to take their eye off the ball when it comes to the face-to-face piece, which it is very clear you need to supplement?

Matthew Oakley: Tony raises a relevant and good point about the risks. We are doing some work at the moment showing how networks and links with friends and family are really essential in finding work and getting people back to the labour market. But there is no reason why Universal Jobmatch cannot form the foundation of that kind of approach where you have an online account; you have a personal account where you might get sent vacancies or ideas. What is missing is the capability of joining that up with the other things that you are doing. Why have we still got the Jobseeker’s Agreement or the Universal Credit one, Claimant Commitment? Why isn’t that there as well, so it says "I have been to talk to five pubs or three shops in person and given them my CV."? Why is that capability not there as well, as a monitoring tool? It is that way of monitoring everything. We would certainly be aware of policy changes, so you see the future of enforcement and monitoring of jobseekers.

Chair: I think we might all be registered with Universal Jobmatch when we get a taster next Monday.

Sheila Gilmore: We will all get catalogue selling.

Q45 Graham Evans: I was in a shop the other day where it said that 50% of colleagues are recruited by recommendation of colleagues, so I think that is true from family and friends, and so forth. Local Enterprise Partnerships (LEPs) have said that the quality and accessibility of the Universal Jobmatch details for the local market data is not as good in comparison to the previous Jobcentre Plus databases. Do you share their concern on that?

Adam Sharples: I have heard that comment but it puzzles me, because I am not quite sure what data they had access to before and what data they have access to now. All I can say is that it always felt to me that Jobcentre Plus was sitting on a mine of useful data on the local economy. Meanwhile, various organisations and people outside Jobcentre Plus were trying to do their own analysis or develop their own understanding of employer thinking, but the two things were not being linked up. For example, I would talk to a Jobcentre Plus and they would know everything about the new vacancies coming into the system, new shopping centres, new employers coming into the area. Half a mile down the road, the local college was struggling to work out what courses it should be laying on and they were trying to guess what employers were looking for, and somehow the two things were not being joined up. All I can say is I would strongly support efforts to make analysis of local labour markets available to other participants who need to understand that.

Q46 Graham Evans: Is DWP aware of this and, if it is aware of it, what is it doing about it?

Tony Wilson: It is a really key question for DWP. Previously, on the website, you could access Jobcentre Plus vacancy data at a local authority level for a whole range of geographies and look at notified vacancies, unfilled vacancies, closed vacancies: the stock of vacancies for all different occupations and sectors. It gave you a partial snapshot of the labour market. That ended in October last year. Since then, you have to go through the Universal Jobmatch system and the data there is not classified by SIC and SOC codes-standard industrial classifications and occupational classifications. LEPs and other partners locally just cannot use it in the same way. It just does not give you the same information. I think DWP is aware of that and alive to that.

Q47 Graham Evans: That is a key mismatch-is it not-into good-quality information about new jobs going in?

Tony Wilson: Absolutely, and there were always issues that colleges, local planners and commissioners had that it was only picking up one part of the labour market, so it was never 100% useful. Nonetheless, it was really important in understanding what, particularly, further education colleges should be providing for people who were unemployed to help them access employment. So we have been developing data tools around this, and we are hoping to do some more work on this. We have some stuff we can show, which tries to use data in different ways. It is Labour Force Survey data about what jobs people move into and what their previous jobs were, but it is nowhere near as rich and useful. There is a really big problem here. I downloaded the Universal Jobmatch top 20 jobs notified not long ago. Number seven on the list of the top 20 vacancies on Universal Jobmatch was oven attendant. Number 19 was diplomat. This is clearly misclassification of jobs, but it is not useful as a planning tool. It was a few months ago when I did that in a quiet moment on a Friday afternoon, but it is symptomatic of a bigger issue.

Q48 Graham Evans: There is lot more work to be done on that, but if you look at the Work Programme support, the Enterprise Clubs looks like quite an interesting one, because you have the Chambers of Commerce in there and various other organisations, which presumably are a good source of what is happening in terms of new jobs.

Tony Wilson: Yes, and I think it is a really good example of where employers and wider civil society organisations and others are doing work off their own backs, as well as through Jobcentre Plus. There is a whole load of really good examples of work clubs that are being run independently of Jobcentre Plus, as well as through them.

Q49 Graham Evans: Detail does need to come into the Universal Jobmatch system, though.

Tony Wilson: Or, as Matt says, it is how you build that bigger picture of what people are doing.

Q50 Teresa Pearce: Tony, you said that Jobcentre Plus can monitor, but they can only monitor if you allow them. You have to give them your personal details to log in. It is my understanding that they do not have a way of just seeing who their clients are and what they are doing. It is personal to each individual. Is that right?

Tony Wilson: It might be worth getting clarification on that from the Department. My understanding is that if you are on JSA that could be a condition.

Q51 Teresa Pearce: They told us yesterday it was voluntary.

Tony Wilson: Okay. That is fine. I missed that.

Q52 Teresa Pearce: To any of you who know the answer, in my local area I am told from time to time there are eight jobseekers for every vacancy. When they say that, are the vacancies the Universal Jobsearch vacancies, do you know? Is that what they are measuring; how many vacancies there are as being how many are on that system? Does anybody know?

Matthew Oakley: That data will probably refer to ONS release of vacancies, which comes out in the labour market stats every month, I would imagine.

Tony Wilson: Historically, often it was the Jobcentre Plus vacancy data, but that was discontinued in October. I do not know what people are using now. It could be the vacancy survey.

Q53 Teresa Pearce: At one time, there were 15 people for every job and then it went down to eight, and I do not know where these jobs all came from that made the difference or if the claimants all disappeared. I just wondered what figure they used.

Matthew Oakley: The half a million vacancies in the economy comes from ONS.

Q54 Chair: Did somebody get those figures by asking employers how much they are looking for? Maybe there are three employers who are bidding for the same contract, so they have quoted the same number of people. They have all said, "We are looking for 100 people," because they will need 100 people if they get the contract, but only one of them will get the contract, so instead of 300 it is only 100 jobs in reality. I know that happens in the oil and gas industry, when we end up sometimes with inflated figures for the number of jobs that will be required in the future.

Tony Wilson: I am sure that must be an issue. For the ONS collecting data for their vacancy survey, which is the most authoritative record of vacancies, I am sure that must be an issue. There are also issues about underreporting of vacancies.

Q55 Chair: Okay, there is underreporting as well, so they might balance each other out.

Tony Wilson: Yes, they may balance out.

Q56 Jane Ellison: Have any of you made an analysis of the extent to which Jobcentre Plus is providing adequate online access, and if you have, can you quantify whether you estimate it to be sufficient for when you see it rolled out? If you have not made that analysis, we can move on to the next question.

Adam Sharples: There is a rather interesting figure that now over half of Jobseeker’s Allowance claims are made online and that number has been rising rapidly: only a couple of years ago it was about 20%. That suggests that, even before the shift over to Universal Credit, there has been a real push on online access.

The only other comment I would make is that we talked earlier about IT tools, digital tools for supporting job search. I just feel that there is so much potential here for doing things more efficiently; doing the benefit claim more efficiently, by collecting the information online and just pushing it through the system and also, secondly, by integrating the different sources of help that can be given to people. For example, if when you signed on, you were given your own account with access to a library of training material, videos, online tools for job search, CV templates, you could imagine a world in which the whole process of job search was so much more helpful, friendly and sophisticated than it is now with your chance to go in and see a Jobcentre Plus adviser for three minutes once a fortnight. Personally, I feel there is huge potential, both on efficiency grounds and because we could be more effective.

Q57 Jane Ellison: I realise that it is obviously quite a big issue, but assuming the right level of support was in place for online access, is there any reason any of you can see that doing more of this stuff online and, in particular, doing more job search online would put off Universal Credit claimants from seeking employment?

Matthew Oakley: We need to go back to the idea that Tony and quite a few people mentioned earlier. I do not see that replacing the need to go in and see an adviser ever, because there has been quite a lot of research in the past from DWP showing that, for instance, signing on by telephone is a bad idea, and increasing the length of periods between signing on times led to slower offflows from benefits. So I do not see it replacing that requirement to go and say what you have been doing to an adviser.

Q58 Jane Ellison: But supplementing it.

Matthew Oakley: But supplementing it, indeed. The adviser can then check the online account with you and say, "Look, you have done these things," and signpost you to a different part of the online account or whatever else that might be. Similarly, if it comes to day 10 and they go to one of the people they have and say, "They do not seem to have done anything," that could just be a quick phone call to say, "You have three days," and that kind of thing, so it is that personalisation as well. Similarly, it could be used as a way of personalising and saying, "We could be pinging you ideas throughout the two weeks." It is that kind of interactive, personalised way. We all know that it is easier to ping people emails than to necessarily give them a phone call all the time, because it is quicker and easier. Again, it is the efficiency idea, so not replacing but certainly complementing.

Q59 Sheila Gilmore: My question was going to be about inwork conditionality. This is initially for Matthew. You suggested a system where working claimants would attend a jobcentre every three months, as one method of doing the inwork conditionality and career progression. Why do you think that would be effective?

Matthew Oakley: Sorry to put this on Tony, but similar to what Tony was saying around the ESA group, this is not about trying to beat people with a stick until they go back to work. This is about saying, "There is support out there for you and as a benefit claimant it should be your responsibility to take on that support where possible." So I am not at all in favour of saying, if someone is working 16 or 20 hours, "We are going to sanction them if they do not increase their hours over the next six months." Clearly, that is not feasible. But what is feasible is, if you are a single male or a couple and you are working 20 hours between you, I do not think it is unreasonable at all to say, "Come into a jobcentre every three months, every six months, speak to an adviser about the opportunities that are available, talk about what actions you might take to try to increase the likelihood of you increasing your hours, that way increasing your earnings and moving more towards supporting your family on your own." The key point to this is the welfare economics and the literature say that people would prefer to earn their own money than take it in benefits, so it is a way of trying to help people to do that, not beat them up for not doing it.

Q60 Sheila Gilmore: How do you get employers engaged in that? Somebody may want to do more hours and many people tell us they do want more hours, but they have to be on offer.

Matthew Oakley: That is the big area where DWP needs to do a lot more work. My view is that the conditionality side of things and perhaps some of the support side of things are much more clear at the individual level, but for the firms it is a lot more difficult. Perhaps it is what we were talking about earlier, around the sectorbased skills academies and that kind of approach, where you are working much more with business in the local area to try to help. This is potentially an area where Jobcentre Plus does less than it might otherwise do. Most offices have one liaison officer for businesses, so perhaps if they are looking to work with people when they are in work, they might employ two or three to build those links. Individuals might move between firms to increase their hours, and that role they were in could be backfilled by a new claimant, so having that flow through to different jobs is important.

Q61 Sheila Gilmore: I have recent experience of quite a large retail employer in the city approaching its workers to go in the opposite direction, to introduce basically zero-hours contracts or a minimum of eight hours if they would not take the zero hours. Some of the labour market forces appear to be moving in a different direction.

Tony Wilson: You can sit that alongside autoenrolment in pensions, for example, where, if employers employ people on low hours, low earnings, they exempt themselves from autoenrolment. There is a nightmare scenario where we are making the benefits system more generous for people working short hours, so we are subsidising work at that end. We are creating a disincentive for employers to employ people above that autoenrolment threshold, because they have to go through the rigmarole of pension enrolment. There is a nightmare scenario where we see a large expansion in the use of zero-hours contracts, short-hours working and so on and that many claimants go along with that, because it suits their personal and familial circumstances and the benefits system is topping up their incomes. There are real challenges about how we address that.

I do not think there are any easy answers at all, but the one thing I would say is that the public sector has a big role to play in this, particularly around social care and the use of zero-hours contracts. But also, we can learn from what good employers do. There are a lot of good examples of good employment practice. Even in traditionally lowpaying sectors with poor progression, there are many that do some really good practice. Typically, that is around effective management structures, personal leadership, effective peer networks and mentoring for people when they are in work, and generally being of a large enough size that you can benefit from economies and turnover and so on.

Businesses often value that kind of engagement as well. The Government probably needs to learn a bit from the consultancy sector about how you do business to business engagement to support businesses to improve their profitability, improve their turnover, improve their progression, and everything else. The CIPD has done work on that. Quite a few different organisations have done work on that and there are even some good examples with Jobcentre Plus through different programmes. We hope to have a report out soon on an ESF programme in Wales, for example, where there are good examples of employer engagement that strays into helping employers grow their workforce, as much as helping them to retain individual members of staff.

Q62 Teresa Pearce: This is probably to Mr Sharples. Jobcentre Plus was an executive agency and now it falls within the DWP. Do you think there has been a consequence to that and, if so, what is it?

Adam Sharples: There are consequences to that. One is, as I was saying earlier, that it is perhaps a little bit more difficult from outside to work out what the structure is, where money is going and who is responsible for what. I am not absolutely sure that Jobcentre Plus exists as more than a brand.

Teresa Pearce: We were there yesterday.

Chair: There were three of them.

Adam Sharples: It is impossible to tell from any information published by the Department, if it exists, how many staff it employs or what its budget is. I see from the NAO’s work that they think it employs 37,000 people. Two years ago, Jobcentre Plus employed round about 75,000 people, so presumably those people who are working in Jobcentre Plus offices but on benefit processing are now being classified as outside Jobcentre Plus. For people outside trying to understand what is going on, all of this does make life a bit more difficult.

Inside the organisation, as Tony was saying earlier, there are advantages. There was probably too much duplication between the Department’s functions and the agency’s functions. But one strength of the agency structure was that there was clearly a business that was focused on its customers. It was clear who was in charge of that business and what its outputs were. There was a good performance framework around that and some pretty strong budgetary controls. I do worry a little bit that some of that could be lost in the new structure, but it is therefore incumbent on the Department to hang on to all of those things and make sure there are clear lines of accountability within the new structure they have put in place.

Q63 Teresa Pearce: Given that everything now is about localism, would there be any benefit to devolving some management to regions, given the variety of job vacancies, different types of workforce? Do you think there would be any benefits in developing management into regions of Jobcentre Plus?

Adam Sharples: The argument for devolving the management of employment support to regions or localities or local authorities is much stronger than the argument for devolving the processing of benefits. It is just in the nature of a benefits system that you need some big clever computers-you need a really efficient process, ideally online-and there is no point having lots of regions or local authorities building those systems.

Q64 Teresa Pearce: Are you saying what would have been the old unemployment benefits office section should be national, but the adviser part and the job search should be more local; that would be better?

Adam Sharples: There is a stronger argument for that. It is quite finely balanced, because the advantage of having a national network is that you can get some consistency, some immediate sharing of good practice, some oversight of the system as a whole. Also, if you are faced, for example, with a downturn in the economy, the national system can respond very quickly and effectively, as we saw in the last recession. It might be more difficult to get an equivalent response at local level. Having a national system also means that, although I know resources are constrained everywhere, at least there is a budget for that system. Once you go locally, you are always faced with trade-offs at local authority level, everything is a single pool and you are competing for resources against other pressures. It is quite finely balanced, but I do agree that, now Jobcentre Plus is essentially the employment side of the activity rather than the benefits side, there is a stronger argument for a bit more localisation.

Q65 Teresa Pearce: Does anybody else have a view on that?

Tony Wilson: Very briefly, Jobcentre Plus is much better integrated now top to bottom within the Department, but there is no clear accountability locally for Jobcentre Plus. There are things they can do without the full localisation of service delivery: for example, cocommissioning of the Flexible Support Funds, saying that they have to engage with local authorities and agree local priorities, potentially cocommissioning of the Work Programme and other support. There are things we can do without the wholesale devolution of Jobcentre Plus about making Jobcentre Plus more accountable within City Deals, within LEPs, with other local partners, which we really should be exploring and testing in the next couple of years.

Matthew Oakley: I completely agree. The City Deals provide a key opportunity to do this kind of idea. What Adam sees as a disadvantage around the local authority budgets and having trade-offs I would say is an advantage. Maybe it is not JCP support that people need; maybe it is housing or other things that the local authority might have responsibility for that would be more effective if they put more money into it. Having that local flexibility means they can put the money where they think it is going to be most effective. Using the City Deals, we can test whether that works at a local scale and then have the evaluation.

Chair: Thank you very much. The bells are ringing, so colleagues are rushing off for PMQs and-I was going to say the Autumn Statement, but this is the Spring Statement or whatever it is-the financial statement. Anyway, thank you very much for coming this morning. We probably could have asked you a lot more questions, but we have covered most of the main areas. Thank you very much for your full replies. What you have given us this morning will help us in our questioning of the other witnesses as well, so that was part of your useful role this morning. Thank you again on behalf of the Committee.

Prepared 18th October 2013