Social Security Committee

Meeting date: Thursday, January 23, 2020

Official Report 652KB pdf

---

Benefit Take-up

Agenda item 2 is the penultimate evidence session in the committee’s inquiry into benefit take-up. We will hear from two panels this morning. We will hear first from a panel on the theme of benefit automation, and then from Department for Work and Pensions officials. I welcome Malcolm Gardner, who is senior business analyst with Team Netsol; Phil Agulnik, who is director of entitledto; and Louise Murphy and Megan Mclean, who are policy and operations analysts with Policy in Practice. Thank you for coming along. We move straight to questions.

The convener reminded me earlier that we are hearing from this panel because of a suggestion that I made previously, so I apologise for that. [Laughter.]

The committee has a number of questions about the practicalities, based on the written submissions, but I am interested in the panel’s view not on the system as it is, but on the system as it could be. In a perfect world, everybody who was entitled to any benefit would be proactively made aware of that. We know that payment protection insurance companies can pursue people to the nth degree and some corporations can access all of someone’s data. If we take the point of view that benefits are an investment in society, that technology could be used to get almost 100 per cent penetration so that everybody who is entitled to benefits gets them. We could maximise take-up of entitlements, rather than people constantly having to jump through hoops or deal with issues because different parts of the benefits system are administered by different Governments. Forgetting about the problems with the current benefits system, can IT systems help us in some way to achieve that?

Shall I take that first?

Yes. I should say to all the witnesses that, if you do not catch my eye clearly, I will not realise that you want to speak, so just let me know.

I will wave.

The answer is that the technology exists, but that is not the difficult bit; the difficult bit is envisaging exactly what you want and what you want the outcomes to be. The issue is much more on the policy side—that is the first thing to understand. Of course, with the general data protection regulation, you need to get individuals’ permission, so you need the citizen to be active in the process and to say that they would like to know about their entitlements. If we consider the way in which companies such as Facebook and Amazon work and how they target, we see that the technology is there—it exists today.

The system would have to be built around the citizen, and the citizen would need to be proactively involved. Some people will not want to be involved, and that will always be the weakness. Also, you are not really Amazon; you are delivering a public service, whereas Amazon is seeking sales and promoting advertising, so it probably worries a little less about permissions. However, you could certainly build such a system—the technology exists today, and it is cheap, really.

I understand the point about the citizen being active, but in many cases the state has the information already. With benefits that derive from a birth, a death or other things that are commonly known, the state has the information already. In many cases, the citizen is not able to be active in the discussion in any event.

There is a fear that the state, in different guises, might be building into its calculations an expectation that many people will not claim because of the complexity. I suppose that my point is about how we get round those issues.

Your question raises a couple of good points. First, what do you do about citizens who cannot engage? In designing anything, you need to have in place a plan B for those who do not engage. Secondly, you asked about joining up data across Government services. It is down to you guys to make that happen, because individual organisations will not want to share data, as information is power. You have to create the framework that says that it will happen.

The objective is quite simple: automation that helps to identify people to be supported, whether or not they are engaged. That is the way of the future. It is down to you to make certain that departments do not work in silos and do not keep information but instead give it up—that is to do with social policy, not technology. The technology can make it happen, but you have to have all the policy in place to achieve it.

I give all the witnesses an opportunity to comment at this stage.

I echo Malcolm Gardner’s comment that the technology is there. As Keith Brown said, the data is already held at various levels of Government; it is about joining that up. Data is already being used to identify fraud, for example, and in some cases benefit data is being shared between the Department for Work and Pensions, local authorities and Social Security Scotland in relation to the carers allowance supplement. When there is a specific, prescribed case, data is already being shared.

Data is also being used in various ways. We take administrative household data sets that are held by local authorities in the form of the single housing benefit extract and data on the council tax reduction scheme. That gives us rich information down to the household level, and those data sets represent millions of pounds-worth of benefit each year that is allocated to households; they give us a lot of information about household characteristics and income. Through that, we can compare what a claimant is awarded with what they are eligible for. As I alluded to, data is already being used to identify fraud. We propose to extend its use to boost benefit take-up; indeed, we are doing that already with various local authorities.

Malcolm Gardner mentioned GDPR. I note that the data-sharing protocol has not significantly changed since GDPR came in, in comparison with the position under the legislation that it supersedes. The biggest change relates to the individual’s right to claim data, which different authorities hold on them. I will give some practical examples. There are legislative bases for sharing data in exactly the way that Keith Brown suggested. Section 131 of the Welfare Reform Act 2012 outlines that data that is held for welfare services purposes can be used and shared for other welfare services purposes, such as boosting take-up. The Social Security (Scotland) Act 2018 takes that a step further. It outlines ministers’ duty to boost benefit take-up among those eligible; indeed, section 7 says explicitly that data should be used, or its use should at least be considered, in discharging that duty.

I will briefly pick up on what Megan Mclean and Malcolm Gardner said. There are two separate issues here. First, there is a vision of a utopian future, with everything being automated and a rethinking of how the main DWP means-tested benefits are worked out. Then there is a separate, shorter-term question about whether some of the devolved Scottish benefits, in particular, could be automated, given that a lot of the eligibility criteria are fairly simple—a person’s eligibility depends on their being eligible for another DWP benefit. If that person has already made an application and shared their data with the DWP, it seems reasonable to expect that data to be shared and their claim to be automated.

The important point about that relates to what a reasonable claimant would expect. If they have gone through a lengthy application process with the DWP, in which they have given it full information about their household and perhaps their disability or their children, they would expect that data to be shared. We should bear that in mind.

I will slightly echo Professor Spicker, who gave evidence at a previous meeting. I do not think that there is a big-bang solution; I think that a series of smaller interventions could be made that, I hope, would add up to things being pushed in the right direction.

09:15  

There are two axes to think about. First, the data analysis that we are speaking about takes people who are entitled to one benefit—in this case, council tax reduction—and looks at their entitlement to other benefits. Non-take-up is about eligible non-recipients, who could claim but do not. There are two varieties of eligible non-recipients; one gets X but not Y. Quite a lot of that goes on. One of the thoughts behind universal credit was that, if everything was lumped in together, people would automatically claim their in-work benefit—the equivalent of working tax credit—with their housing benefit and we could get rid of the partial take-up problem. However, as Professor Spicker pointed out at length, the flip side is that there is a big, difficult-to-manage benefit and, when it goes wrong, everything goes wrong in the household. I will not carry on about that.

One problem is that people are getting something, but are they getting everything? Benefit calculators have a lot to do with that. Another problem is that people do not know that they have an underlying entitlement to something. Typically, they are people who are in work and who are often tenants who have not realised that they had an underlying entitlement to housing benefit under the old system or have such an entitlement with universal credit. Another affected group are better-off pensioners, who are often entitled to council tax reduction, which goes quite a long way up the income scale, but who do not realise that. That is the first axis, and it is about whether people are getting everything or whether take-up is partial.

The other axis is about people who are going through the system—for example, someone who has just had a baby and wants to claim everything. People can interact with them. If the user journey is online or on the phone, people can ask them extra questions, take permissions and say where they will send the data, and the approach is relatively uncomplicated. The complicated bit, with GDPR and everything else, relates to people who are in the stock of the system. Trying to get into the stock of the system to find out who those people are and proactively contact them in some way involves big and difficult information technology projects.

That is really interesting, because there were different answers. The first point that Mr Gardner made was that the technology exists and that automation can be done. However, we immediately revert back to complications such as GDPR. We could look at incremental benefits, but in my view incrementalism has not worked over the decades—there are still appallingly low levels of take-up.

Somebody talked about a utopian vision, but I think that we can hold out and say to state actors, “This can be done.” I had in mind the idea of people, through smart technology, getting a prompt that is based on the information that the state already has about their date of birth, their national insurance number and the fact that they have had a baby. It is also about not just the two Governments but individual local authorities and others that provide benefits. As we have heard, there is no reason why the technology could not support that approach. One would think that, if we were all signed up to the idea that we wanted to maximise people’s benefits, as doing that would be a good thing, holding out that possibility would start to change attitudes towards GDPR and the different benefit systems.

That is pretty much how we would describe the next generation of our BECS claim form. Doing things in that way is exactly what we are looking at.

What I took from Mr Gardner’s responses is that automation can be done if there is political will. I think that Megan Mclean or Louise Murphy pointed out that Social Security Scotland now has a statutory duty to have a benefits uptake strategy. However, I am not sure that the DWP and Social Security Scotland have a statutory duty to inform individuals about benefit entitlement when they become aware of that. Local authorities could be aware that an individual or a household might have an entitlement, but they are barred from putting the data together and informing the individual because of GDPR. The same could apply to the DWP or Social Security Scotland.

Would a statutory duty to inform perhaps circumvent some of the GDPR issues? Is it just a case of getting the Information Commissioner in the room with the legislators and asking, “How do we fix this policy outcome?” Surely it would be much more straightforward if we took an ask-once approach to individuals who want their data to be shared on the basis that they could maximise their income.

What you have said is absolutely true for people who are going through the system—for example, those who have just had a child. Permissions can be taken from them as they do that. The problem relates much more to people who have an existing claim and do not have a change. Those people do not interact with the system, so we need to go in and get their data. However, I look to other people on the panel to explain the GDPR rules for doing that.

If there is a new claim, all the consents and so on can be dealt with in real time—on the phone, online or wherever. The problem comes when we do not interact with the person.

That is very helpful. If there is a static situation with a claimant, opportunities might be missed, but an additional opportunity is presented when there is interaction. I think that that is what you are saying.

If there are consents to be taken—for example, if you want to send a person’s data to the Scottish system—it is easier, because that can be done there and then. However, if you can see that an individual has an entitlement but they are not in contact with you, you just have that information from the stock of data. That is a different situation. I cannot comment on that, because I do not know the rules.

I probably should not explore the issue further, but I will let Mr Gardner back in.

We should remember that GDPR is about permissions; it is not about blockers. We should not be saying that, because GDPR is silent on a subject, it does not allow it. That is not the way that we should look at GDPR. GDPR simply requires that we think these things through and justify what we are doing, through either a legislative change or additional legislation. GDPR is not there to stop us doing this stuff, especially if doing it is in the public interest.

Other witnesses might want to come in, but I had better push on with other questions.

The committee has been looking at automation, because of our ambition to drive up entitlement uptake. Some of the written evidence that we have received has pointed to areas in which automation will not necessarily be a good thing and could impact on uptake. I want to explore both sides: the positive and negative impacts of automation.

Which types of automation would be beneficial to increasing uptake? Do you have any examples of where such automation could be introduced almost immediately or in the relatively short term?

On your last point, automation is happening. Many local authorities have automated council tax support claims. If they get a notice from the DWP that one of their residents is claiming universal credit, they will often take that as a notice to start the council tax reduction claim. That is a good example of automation being fairly simple. Often, local authorities have the initial data, which maps on to their council tax support eligibility criteria, so they can get the ball rolling. Looking at that example, we can see that automation is not something that will happen in the distant future; it can happen fairly soon, and it is already happening in some cases.

Mark Griffin asked about the positive and detrimental aspects of automation. We need to think about what the customer would think. It makes people’s lives easier if they do not have to share information again and again. Their trust can be diminished if they think that there is no joined-up working between authorities.

In relation to the detrimental aspects, we want to ensure that there is still human interaction when that is needed and that vulnerable claimants who might not be able to use an online claim form, for example, still have the face-to-face support that they require. Automation and technology need to fit within a wider benefits system that still involves face-to-face interactions.

I echo some of what Louise Murphy said. We are proponents of data being used alongside clear communication. We would not necessarily suggest a system whereby a claimant receives differing award levels without any explanation as to why. Given the onus on claimants to monitor their own claim, we need to communicate clearly to them why their awards might change from one month to another—for example, their eligibility might have been seen to be relevant in another area. If we do not make that clear, a claimant might be unduly concerned, particularly in a culture in which, if an overpayment of a benefit is awarded to a claimant, the overpayment is deducted in the future, regardless of whether the claimant knew about it.

As Louise Murphy said, we are saying that data is not the only answer and that automation needs to go alongside clear and respectful communication.

Phil, did you have something to add?

My examples are all about people who make claims when they are in the system. Calculators try to point people to their other entitlements. We do not do a lot of claim forms, although we do a council tax reduction claim form. Our ambition there is for the data that we collect that is relevant to the Scottish social security system—because social security benefits are passported, that is really just someone’s name and address—to be able to just flow across into the form.

The idea of having separate forms for separate benefits is very much part of the social security system. The existence of lots of different agencies is fixed into the system, but those agencies should share data so that, if someone fills in one form and then has to go to Social Security Scotland to do their Scottish child payment form and their council tax reduction form, they do not have to re-enter all that data. That is obviously a good idea and that is relatively simple to do, because it all happens in real time and it is possible just to take the permission.

That is an example of how automation can make things simpler for people who are going through the system and reduce the need for retyping. I do not have any examples of automation that involve going into the stock of the data, which is a lot more difficult.

I want to pick up two points. As has been mentioned, universal credit provides a benefit as a result of the fact that benefits are wrapped up together. With universal credit, there is a challenge and perhaps an opportunity to do with the fact that, as has been inferred, the data that is held by the DWP is a really rich data set that covers what was previously six different types of benefits. If that data were shared with local authorities, that would give them a powerful case for doing things, and not just in circumstances in which a claimant’s situation had changed. If councils had that data, they could dive into it and use it to look at stagnant claims that had gone unnoticed for a while, as well as to compare eligibility. We have been working with local authorities to bring out—using their housing benefit and CTRS data—stagnant claims where eligibility is not fully understood.

I reiterate that we strongly believe that, as universal credit rolls out, the real power will be held within that data set. There are cases of the DWP sharing certain elements of UC data when there is a clear, prescribed case for its doing so, but we would like the DWP to share it more routinely with local authorities in the future, because that would allow local authorities to engage in the notifying or automation of take-up that the committee is here to discuss.

Do you want to explore that further, Mark?

No, thanks.

Do you want to come in on that, Shona?

Yes, thanks. I want to follow up on Megan Mclean’s point about discussions with councils.

I take it that you are saying that the data that is being looked at is more on the local authority side. It is interesting that there has been more interaction on fraud than on boosting benefits—we might come back to that with the second panel. Is that process focused mainly on things such as council tax reduction? What other areas are local authorities looking at? How universal is that process? Is it something that all local authorities, or only some of them, are doing? How many of them are doing it?

We are using council tax reduction data along with housing benefit data, but, because of the richness of the data that is held in those two data sets, we are working with local authorities to target support at the take-up of pension credit, for example. We are also looking at the severe disability premium within means-tested benefits, whether that is housing benefit or a DWP-administered means-tested benefit.

We are also looking at more discretionary areas such as discretionary housing payments, and we are even looking at universal credit and jobseekers allowance. The approach is being used primarily around means-tested benefits. In certain areas, automation could not be the answer. For example, with disability benefits, some human interaction is needed and assessment could never be automated. However, we are working with a number of local authorities on the means-tested benefits that are related to income.

09:30  

On whether this work is universal, the answer is no, for many of the reasons that we have already discussed. There is a perception that the GDPR is a blocker, as Malcolm Gardner said, although the perception of that varies quite widely between the local authorities that we have worked with. The place where we have seen the greatest progression past such blockers is the project that we are working on with the Welsh Government to analyse the impact of universal credit on council tax arrears, for which we are using the same data sets across all the local authorities in Wales.

Initially, we ran into some blockers when we were trying to collect that data, because some local authorities were hesitant about passing over the data and signing the data sharing agreement. I support what the convener said earlier about guidelines and the legislation perhaps needing to be a bit tighter, because it was when we had the Welsh Government produce guidelines for local authorities that we saw a change in perception. Once the centrally-provided guidelines were sent round, we were able to gather data from all 22 local authorities in Wales.

That is very interesting. How many Scottish local authorities are taking part?

Currently, we do not have the data from any local authorities in Scotland, although we are in discussions with a few—discussions have happened this week, for example. Glasgow City Council has a scheme whereby it uses the same data to administer the school clothing grant. A passported benefit that provides eligibility for that grant is council tax reduction, which makes it quite a straightforward application of the use of data to identify need. Glasgow City Council has not moved beyond that scheme to include the more means-tested benefits but, as we have said, the data and the technology are available to all local authorities.

Do any of the other witnesses want to add to that?

I have a couple of points to make on that. First, we are working with three local authorities in Scotland in that very area to look at how we can build the technology and create that sharing. However, we come at it from a slightly different angle. We believe that the local authorities should do it for themselves, so we have built the tools around getting them to do the work and facilitate it.

Secondly, although we are not doing much at present around Scottish social welfare, we are working with a number of English local authorities that have responsibility for the equivalent benefits. We are running with about eight local authority systems that allow people to come in and say that they need additional help, and we are looking specifically at the discretionary end rather than at the means-tested end of producing whatever help they need. Basically, the system identifies whether they need a DHP, assistance with getting to school, retraining or help with budgeting—almost anything that the local authority will consider. The questions are asked just once, and the system identifies what the person needs and directs them down the shortest route. It is an automated process of identifying what they need.

We are also working with an artificial intelligence company in Manchester and considering how we can use things such as chatbots and native language alongside AI machine learning to speed up the process even more and automate it. The technology is there, as I said earlier. It is a case of working out how to make it work, and that requires working with the local authorities and the public.

Before I bring in other committee members, are there any other comments from the witnesses?

As we heard, only certain local authorities are currently working with companies such as ours to do full take-up work with the data. However, at the moment, every local authority throughout the country has access to these rich data sets—the housing benefit data set and the council tax support data set. Even if local authorities are not doing any on-going, complex work on that data, they have access to it and they can work out, for example, discretionary housing payment eligibility using it.

It is really important to bear in mind that, as universal credit rolls out, there will be less and less visibility on residents’ data for local authorities. A lot of the proactive take-up work that local authorities are doing will become much harder if we do not get a commitment that universal credit data will be shared in the same way.

For the record, I will read out something in your submission that, for me, is quite important in all this. You say:

“One of the biggest blockers to take-up is data protectionism, whereby data owners create pseudo data protection rules that places blockers on the appropriate sharing of data. A recent report ... ‘Customer Experiences of Tax Credit Overpayments’ highlights that customers are unclear about what details they have to report and to who.”

That strikes me as being absolutely relevant to what we might all be trying to achieve here. It has been my experience as an MSP that, since the data protection legislation came into force, there has been confusion over its purpose. I had a ridiculous situation yesterday in relation to my inquiring at a local public college about where a student’s bursary was in the system—I did not want any details. In the old days, before the new legislation came into force, you would be copied in without any consent form, although the national health service has always been different. We need first to sort out the confusion about data protection and what the Data Protection Act 2018 was actually for—do you agree?

Was that a quotation from Mr Agulnik’s submission?

I do not know anything about GDPR, I am afraid.

Was it from Mr Gardner’s submission?

Yes.

I will ask Mr Gardner to answer first, then, to give the other witnesses an opportunity to think about the question.

I have fairly long experience of data protection, all the way back to its original form. We spoke earlier about using the technology for finding fraud within systems. I have to put my hands up and say that I am the father of data matching for the purposes of fraud investigation, which we started in 1992 and which has since been much copied by people such as the Audit Commission.

We had a real handle on the use of data, but, with all this data sharing and with data sharing in general—even on the level that Pauline McNeill spoke about, when she was just asking a question—what now tends to be said is, “Data protection won’t allow this,” or “GDPR won’t allow it.” Actually, there is nothing in any of the legislation that says that people cannot do something. The legislation says that, if the data is required for one of the purposes that it lists, you can do something with the data.

I recall that, when council tax was introduced, it took three years to get the legislation that allowed for the sharing of that information, because it was covered by secondary regulations. The only thing that data protection stopped was the sharing of that data until those regulations came in. There is nothing in data protection to prevent such sharing, but people will just make it up, making it convoluted and difficult.

The question should not be, “What will prevent me from sharing this data?”. The question should be, “Is it in the public interest to share this data and how do we go about doing it?” That is how it should work.

Malcolm Gardner is completely right. There is confusion about data protection, as there is nothing in GDPR that prevents any of the data sharing or processing that we are talking about. As Megan Mclean said, the Welfare Reform Act 2012 still stands—it has not been superseded by GDPR—and it explicitly allows for that sort of data sharing.

Megan also spoke about the Welsh Government, and we require a bit of clarity. The purpose of GDPR is about making all the data protection laws more modern and harmonising them, which is a good thing. It is also about allowing people to know what information is held about them, and—if they want to—to access it, which is sensible. We need to go back to those principles and try to prevent any further confusion.

Further to that, I want to examine how we might push that forward through what the committee could do in the form of recommendations.

You highlighted Glasgow; I have experienced Glasgow City Council’s brilliant attempts. For example, a cheque was sent in the post to families who the local authority was able to identify were entitled to the school uniform grant as a passported benefit. However, some local authorities are now concerned that data protection will get in the way of such action.

As to mechanisms for resolving that, what if there was one big consent form to be used among agencies? I accept that such a form would target those who already interact with the system, and that you would have less chance with those who do not interact with the system. Nonetheless, you might want to focus on and deal first with those who interact with the system. It could be that people who arrive at the agency would get one big consent form that says that the person agrees to share their data for the purposes of benefits take-up. Would that make a difference?

The answer is yes. That is, fundamentally, where to start. When people sign up for Facebook, Google, online banking and so on, very few of them read the terms and conditions. However, they are largely signing up to allowing their data to be used for a load of stuff. I am not suggesting that it should be done in secret or be hidden: people need clarity. It is simply about telling people that you will use their data for specific purposes that are in their interests. That is what it comes down to.

I agree with much of that. I will mention some of the issues that we know of around universal credit. The universal credit application asks specifically whether the claimant would like to take up council tax reduction. We know that some local authorities take that as a means to notify the claimant that they need to make a separate claim for CTR. Others simply run the information of a UC notice all the way through their software to award the benefit. In local authority areas where that does not happen, we see a drop-off of about 10 per cent, and up to 15 per cent, in the number of claimants who take up council tax reduction when they move over to universal credit, despite the fact that they are still eligible for it. Some of that is to do with confusion; if a person answers a question about CTR in their application, it is reasonable for them to expect that that will be taken up by someone else.

On a separate point, some of our clients have launched take-up campaigns and have followed them up with claimants. That relates to what Malcolm Gardner said about whether people would be happy to have their data used in that way. One of our local authorities—Luton Borough Council—reached out to 88 households about maximising their income through benefit take-up. Of those 88 households, 87 said that they were actively happy with the fact that that had been done, and one had a question related to confusion about how the process had been launched. The authority received no negative feedback. Although, of course, it is necessary to take into consideration the fact that it might not be everyone’s view, it is, by and large, reasonable to think that people will be happy for their information to be used for their own good—that is, to maximise their income.

So, the committee could—if we chose to do so—write to every local authority in Scotland asking whether they ask for that for universal credit claimants?

Each local authority receives a notice from the DWP when someone in its cohort has moved on to universal credit. The committee could ask local authorities whether they are making use of that data. Depending on its software, the local authority might be able just to run the data through its system and award the benefit. If it does not have software that would enable that, it could just notify the claimant. The data enables either of those things to happen. That would boost take-up of CTR, which can be quite a significant portion of income; in some cases, it represents up to a third of a household’s income. Clearly, it is a meaningful benefit and a good target in terms of income maximisation.

09:45  

Would it make sense to focus on one benefit, such as council tax reduction, and go through the processes to see how it could be better automated, so that there was increased uptake? Is that benefit a good one to consider in that regard?

Yes. There are cases in which that has been done. I am happy to put the committee in touch with local authorities that are already doing such work.

However, although work can be done on certain cases in the short term, our view is that there should be greater linkage throughout the system so that we do not need to run short-term campaigns to boost certain types of income. We hope that it will one day be the case that most people are covered at all times, and that there is clarity around eligibility, and a drive for take-up of each element of the system.

Do other witnesses want to comment?

One issue is about local authorities accepting data that is produced as a result of the DWP asking whether a person wants to claim CTR. Many of the problems with the system arise from the fact that we do not know where people are coming into it. However, the typical model is that people claim a DWP benefit, from which a council tax reduction is almost passported. That flows from the idea that, in the legacy benefits, the person would claim jobseekers allowance when they lost their job and would then go to the council to claim housing benefit and council tax reduction. Typically, things were done that way around. However, they do not have to be that way around.

Some people will qualify for only a council tax reduction and will claim that, so there is a different kind of automation that does not involve the DWP—in general, that is a good thing. For example, someone who is filling in a council tax reduction form for Glasgow, or Edinburgh or South Lanarkshire could be asked whether they have a child under six and, if so, whether they have claimed their Scottish child payment. That would be from August onwards, of course. If they would like to claim that benefit, their data could then be sent to Social Security Scotland. That is a lighter-touch approach that involves Scottish councils talking to the Scottish Government and it is done through real-time interactions with individuals, so permissions can be taken there and then. When the process involves the DWP, we have to get deep into the data and it becomes a project, so we end up with individual local authorities running projects for this and projects for that.

We must, however, get people’s permission. I assume that people would rather have money than not, so if there is something that has all the evidence to show that the person qualifies for the Scottish child payment, we might ask whether there is an underlying responsibility to share that with Social Security Scotland so that it can start paying that benefit to the person. We know—as crazy as it sounds—that signposting vulnerable households does not mean that they will proactively go and seek the benefit, even when there is money at the end of it.

In terms of flowing through the system quickly, I assume that people are applying online—if we are talking about people with young children, that is a reasonable assumption, because online usage is higher among that demographic.

The information that one enters on a form on, say, South Lanarkshire Council’s website is owned by the council. Our proposal would mean that, when one goes to the Social Security Scotland website, the form there will be pre-filled and, when the claimant clicks on the button to submit the information, they will be told that the data will sit with Social Security Scotland. There are two sets of data, but by pre-filling forms, we could make it easy to give the two sets of data to the two agencies, which would speed up the process of claiming. Does that make sense?

Yes, it does. Following Megan Mclean, we will move to the next line of questioning.

The convener mentioned that even people who are identified as being vulnerable and are eligible might not follow up claims. That, again, comes down to the partnership of data and technology and human contact. Part of the value in using data and technology is that they provide visibility of who needs additional support.

Many people are able to fill out an online form and, indeed, are willing to do so. However, when it is identified that a person is more hesitant about doing that—it could be to do with their online capabilities or something else—the role of an adviser or the local authority is all the more important in engaging the household, figuring out why the person is hesitant and identifying what can be done about that.

Mr Gardner has indicated that it is down to us guys to find solutions, and that there are policy solutions. The witnesses have mentioned one or two ideas for policy solutions. Where is the balance in respect of policy solutions that lie in the hands of this Parliament, and the responsibilities that lie in the hands of UK Government and agencies, on issues such as data sharing?

That is a tricky one. Westminster probably has the big guns at the moment. I would probably draw on some of the stuff that Phil Agulnik has mentioned. Quite a number of welfare benefits are coming to Scotland: those are the ones on which the committee should concentrate. Unless you can exert major influence on data sharing—as far as I can see, virtually no one has any major influence on the DWP on data sharing—you need to make what stuff you have work as smoothly as possible and not just duplicate some of the thinking that has gone before. Part of the problem is that people always naturally say, “This is how it’s worked before—let’s do it that way.” You need to think about things differently.

New technology should not be allowed to drive the solutions. Universal credit was much more about having a system in which people could claim online than it was about the underlying legislation. It should have been the other way around—it should have been first about the underlying legislation, before the question how to deliver on that was asked.

That is where I am coming from. Parliament should think through policies and about what it wants to achieve. After that, ask about the technology that is needed in order to deliver on that. Clearly, that includes focusing on the general data protection regulation and asking whether the system is to be permissive. If the it is not, you will need to think about what to change to make it work.

On policy versus how things work in practice, when the committee is thinking about future devolved benefits, it should bear in mind the eligibility criteria. A lot of what we have been saying about automation relies on the assumption that information that we are sharing maps to the eligibility criteria—that is, the data would allow us to administer the benefit. However, some benefits, especially disability and caring benefits, require much more nuanced information.

The criteria should be as simple as possible, so that as much information as possible that is already held can be used. The committee received a submission about Glasgow City Council automating the school clothing grant. Eligibility for that benefit depends on council tax support eligibility, and the council has that information. However, the council has said that it faces a barrier: it cannot automate free school meals because eligibility for them has additional criteria. Therefore, in finalising the details of benefits, criteria must be as simple as possible.

If I have picked you up right, you are saying that the design of the criteria has as much to do with the simplicity of a benefit as any operation or form.

Yes—exactly. For a benefit that has simple criteria, sharing information is much easier. If a benefit relies on data that comes from many sources, or if the criteria are different for local benefits, devolved benefits and DWP benefits, it is obviously much harder to get the information.

We heard earlier that more than 16,000 households in Wales that were eligible for a severe disability premium were unaware of that. The severe disability premium is a DWP benefit, so it is perhaps surprising that those people have not picked up on that entitlement and the money has gone unclaimed. Is it fair to say that the DWP could do some of the work that we are talking about but is not doing it?

You have identified one of the major barriers, which is cultural. Even within one organisation—the DWP—data sharing and proactive use of data are not happening. The severe disability premium is difficult because it can be missed because of how data is reported. For example, when a person is living with an adult child or another adult who is not their partner, that prevents their being eligible for the severe disability premium, but that might not be picked up on a claim form. Also, a change of circumstances might affect eligibility but might not be reported. The situation is complicated; I agree that a lot more could be done. We have talked about sharing between departments, but in the first instance, a lot of work needs to be done just within departments.

You said that there is an issue about the cultural attitude to information, and that there is not a lot of proactivity when it comes to using it to ensure that people get what they are entitled to. Perhaps that is reflected in the fact that there is so much unclaimed entitlement.

Yes. We often find that local authorities are proactive because, with stretched budgets, it pays for them to do that. Especially in homelessness reduction, it pays local authorities to use the data that is available to them to identify people who are at risk of homelessness before they become homeless. That is an example in which a culture shift is gradually happening because local authorities are thinking about overall returns on investment. We have done a lot of work on that. We realise that there will be an up-front cost for use of data in certain ways, but given the huge costs of temporary accommodation, for example, identifying such people is a useful way to spend our time and money.

It is about investing to prevent greater costs and distress as time goes on.

We are undertaking the inquiry, and great work is going on across Scotland. However, a lot of it is being done with groups of people who already claim something but might not be claiming everything to which they are entitled. Policy in Practice’s written submission mentions the particular challenge with households who are entitled but who have no or limited contact with welfare rights services, Social Security Scotland or the DWP. Do you have any idea how big that group is and how we might better reach out to those people?

For certain benefits, there is clear data on uptake. For example, the DWP’s data shows that only about 60 per cent of people who are eligible for pension credit take it up, which happens for reasons that Louise Murphy mentioned earlier.

The ease with which data on take-up can be collected depends on the criteria. There is more difficulty with benefits such as employment support allowance and disability living allowance, for which the criteria might not be as visible outside the benefit applications. There is clear information on some benefits and less on others.

DWP analysis estimates that more than £10 billion of income-related benefits go unclaimed in the UK. We also know that there is a rural-urban divide on uptake of benefits: typically, uptake is lower in rural areas. Given that much of Scotland’s population is in rural areas, that is something to consider when thinking about the challenges in collecting data.

That issue of rurality impacts greatly on Scotland. Based on your experience with the DWP and everything that it is doing, or not doing, to increase take-up, would you say that Social Security Scotland has an opportunity to learn from that and address such issues?

10:00  

Yes. Social Security Scotland has a great opportunity, and we can see that it is trying to get the infrastructure in place by setting up offices throughout Scotland.

There are two issues with take-up in rural areas. There is an issue of infrastructure and whether people have access to jobcentres or advisers—that is more of an issue for rural populations than for those in urban areas. There is also the issue of communication and clarity. Social Security Scotland has a great opportunity to work with data and to put in place the correct infrastructure to plug those gaps. We encourage it to work with local authorities to use the data that they hold and, if it can, to encourage the DWP to share data. There are various opportunities at different levels of government. Wherever possible, we encourage data sharing, clear communication with claimants and putting in place infrastructure that allows people to easily access services.

I am going to widen that out and ask whether the other witnesses want to comment, but first I want to give a time check. We have the DWP giving evidence immediately after this evidence session, and we have about 10 minutes left. Mr Balfour has been very patient and still has a line of questioning that he wishes to pursue. I will bring in Malcolm Gardner and Phil Agulnik on Alison Johnstone’s point and I will then see how much time we have left for any follow-up questions.

I want to pick up on the point about the difference between urban and rural areas. There are a couple of key points on that. One is that, with Social Security Scotland, there is an opportunity to build something new, with a new relationship and a new contract, and that is important. With all due respect to the DWP, I think that universal credit has damaged trust between the public and the department, although I will not dwell on that.

Despite my accent, I am actually from Oban, in Argyll and Bute, so I know that that rural area is different from rural Wales, for example. There are a lot of cultural and community issues that need to be dealt with. I go back to Megan Mclean’s point that communication is incredibly important. In order to increase take-up in some of those areas, much more is needed than just technology. The approach needs to be built around communication and trust. That is important but, as we are short on time, I will not say any more.

I appreciate that.

Briefly, the problem with people who are not in the system at all is the bigger problem. Across all benefits, including Her Majesty’s Revenue and Customs, the total unclaimed is about £20 billion, and the extra bit in universal credit from increasing take-up is claimed to be about £2 billion. Therefore, the problem with people not being in the system at all is the bigger one.

A good example of that is better-off pensioners who have an underlying entitlement to pension credit. There will be a good opportunity for that group when the television licence becomes means tested. That kind of classic communication event when people are told, “You may not have to pay—do you know whether you are entitled?” could be effective with that over-75 client group. Maybe come the autumn and Christmas, there will be a similar effect with better-off tenants with children who qualify for the Scottish child payment. They might have a small underlying entitlement to universal credit or child tax credit, but they will get a boost from the child payment. Events such as new benefits being introduced or changes to the TV licence are good opportunities to increase take-up.

Alison, do you have a follow-up question on that point? It would have to be brief, I am afraid.

We are talking about issues of rurality, such as the fact that a typical Highland village will not have a jobcentre. Could we do more to take jobcentres to people and to have welfare rights advisers embedded in places such as general practitioners’ surgeries? We know that such things are already happening, but do we need to do more to make them the norm?

We could do that simply by making the most of what is already there. For example, housing associations often have good relationships with clients. Many of them also have welfare rights advisers, as do other organisations such as disability charities. One benefit of the universal credit system is that it encourages a claimant to ask, “Who knows me best?” We could aim to have jobcentres throughout the country, but that would be a large and costly undertaking. Instead, perhaps we should look at what is already there and see whether we could upskill or train staff with whom people already have the relationship of trust that Malcolm Gardner has rightly emphasised.

I thank our witnesses for coming. As something of a Luddite, I find this area interesting but slightly confusing. In Scotland we have a new social security agency that has just been established, and we are on a journey that will take us towards having powers over a number of benefits. If I might put it in simple terms, what is the one thing that the new agency could do now that would make it easier for people in Scotland to take up the devolved benefits? You have already dealt with a number of aspects of that in response to the questions of Alasdair Allan, Alison Johnstone and the deputy convener, but what would be the main element of a culture that would encourage take-up in Scotland?

That is a very good final question, but do you have any others that you would like answered first?

No.

I am tempted not to come back for further questions after that one has been answered, because we will have to move on, so this is a good opportunity for our witnesses to put on the record anything that they would really like to see happening.

If I might be cheeky, I would like to suggest two things. Phil Agulnik has rightly made a distinction between those who are underclaiming but who have active claims and those who have no claims at all. Among the first group, where we are trying to boost take-up among those who already receive at least one type of benefit, the approach should be to use the data that is available. A rich data set is available through local authorities and perhaps the same will be true of the DWP in future.

However, for those who are not claiming any of the benefits for which they are eligible, the approach will come down to communication and creating the culture of trust that was touched on earlier. Social Security Scotland put it quite well when it said that it needs to build dignity, fairness and trust in the system. Therefore we need to see clear communication and an infrastructure that allows partnership, forms a network that supports people to better understand their eligibility and empowers them to make decisions within that.

My one wish would be for there to be a focus on the benefits that can easily be automated, such as the passported benefits that are being devolved to the Scottish social security system. We could use the information that we have from our experience of automating the carers allowance supplement to automate other benefits. That would create a consistency in the system that would mean that information could be easily communicated and understood by people. To emphasise again the role of trust in communication, if one devolved benefit was paid automatically and another was not, that would just create confusion. People would think, “Why are they doing it in one area and not another?” Therefore where it is possible and simple to do so, we should focus on automation.

At the risk of repeating myself, I will say that the main aim should be to focus on the agency’s relationship with Scottish local authorities. A claim for a Scottish local authority benefit should be viewed as a claim to the Scottish social security agency—the two should be aligned in people’s minds. I do not think that that could be done with the DWP. Trying to get the things that you are after from it would involve a long journey that might or might not succeed, whereas working with Scottish local authorities is meat and drink to the Scottish Government. The ambition should be to have a social security system that is seamless within Scotland. Achieving one that is seamless in conjunction with the system in the United Kingdom is probably too big an ask—at least for now.

I agree with everything that has been said. Those are all relevant issues, and I will try to address the simple point that underlies them. The new agency must not only want to help people but be seen to be doing so—it must make those people feel that they want to go to it for help.

Thank you, Mr Gardner, for that final comment from our witnesses. It is worth putting on the public record that, yesterday, members had the opportunity to visit Social Security Scotland, where we had very positive and constructive discussions with its representatives.

I thank all four of our witnesses for giving us their time. I ask them to follow the committee’s inquiry. Our final evidence session will take place next week, when we will hear from the cabinet secretary, and we will publish our report a few weeks after that.

I suspend the meeting briefly before we hear from our next panel of witnesses, who are from the DWP.

10:10 Meeting suspended.  

10:14 On resuming—  

Welcome back, everyone. I welcome the following officials from the Department of Work and Pensions: Katie Farrington, director for universal credit and employment policy; Andrew Latto, deputy director for devolution, pensioner benefits and carers allowance policy; and Derek Kilday, group director for central and west Scotland. I thank all three of you for coming along this morning. I know that you sat through the first evidence session—thank you for taking the time to do that as well.

I invite Derek Kilday to make an opening statement before we move to questions.

Good morning, and thank you for inviting us to today’s meeting of the Social Security Committee. As we will be discussing benefit take-up issues in more detail over the next hour or so, I will not go into specific points in relation to that now. However, I thought that it may be helpful to the committee if I made some general points by way of an introduction.

First, on behalf of the DWP, we welcome the committee’s interest in the work of both the UK and Scottish Governments on the take-up of benefits. Although our evidence will focus on benefits that are reserved, their interactions with the Scottish Government’s new and existing benefits are a key element of the services that the two Governments provide to the people of Scotland.

The benefits in the reserved benefits system divide broadly between those for people who are above the state pension age and those for people below it. In both groups, personal circumstances vary. My colleague Andrew Latto will explain how we will contact people as they approach the state pension age and about pension credit, which tops up the state pension for those who are eligible.

People who are below the state pension age and who make new benefit claims tend to come to us through universal credit, and we work closely with a wide range of organisations to provide support. For example, the agreement that we have with Citizens Advice Scotland provides support under help to claim. The committee will have seen evidence of a wide range of initiatives to increase take-up from similar organisations. You also spoke this morning about the role of local authorities, and we acknowledge their valuable work.

The DWP offers advice over the telephone or face to face, and we can make special arrangements to assist vulnerable people to access our services. Our own work coaches also offer tailored help at a local level via the Jobcentre Plus network across Scotland. The DWP also offers benefits for carers and for people with additional costs that arise from disabilities or health conditions. Those are now, largely, matters for the Scottish Government. We have already worked closely with it on the carers allowance supplement, and we will continue to work with colleagues at the Scottish Government to provide a smooth transition in areas such as the personal independence payment and disability living allowance. We are acutely aware that, as further benefits become devolved, people in Scotland will be receiving benefits from both Governments, and it is incumbent on us both to ensure that the customer receives the best service possible.

In the future, many pensioners will receive a state pension from the DWP and winter heating allowance from the Scottish Government. Some will also receive disability allowance from the Scottish Government and pension credit from the DWP. Many people with disabilities and health conditions will receive universal credit or employment support allowance and disability assistance from the Scottish Government. Many carers will receive carers assistance from the Scottish Government, topped up by universal credit. As the committee can see, there is a lot of interaction between the Governments, and, as a department, we are committed to ensuring a smooth transition.

We recognise that the social security system provides crucial support for many people. It is important that access to our services is as straightforward as possible and that people do not face unnecessary obstacles in applying for benefits to which they are entitled. We look forward to answering the committee’s questions on how we can achieve that.

Thank you very much. That was very helpful. I will kick off, and I will pick pension credit as an example.

It is a really helpful example, because it is one of those benefits that has a relatively low take-up of only 62 per cent. That means that up to 38 per cent of people who would be entitled to pension credit are not receiving it. In previous evidence sessions, we heard that it is sometimes difficult to work out who those people who are not receiving it are, let alone to reach out to them to encourage them to take up that entitlement. Does the DWP or the UK Government have data that tells you who makes up the estimated 38 per cent of the cohort of pensioners who are entitled to pension credit but do not take it up?

The short answer to that question is no—we do not know the households. We have administrative data that tells us who is claiming, and we compare that with the household survey of about 20,000 households, whereby people report their capital and income. The difficulty with pension credit is that—unlike for the state pension, for which we look only at a person’s date of birth and national insurance record—we also need to know what is in a person’s bank account and what their income is. We know that from survey results at an aggregate level, but we do not know it at an individual household level.

That said, in 2010, we conducted a pilot jointly with Age UK in which we used the data that we hold, that local authorities hold and that HMRC holds to get to a proxy of who we thought would be entitled on a sample basis. We paid people pension credit for a three-month period, and, at the end of that period, we invited them to claim. Rather disappointingly, only 9 per cent of the people who had been receiving that amount of money claimed it at the end of the process.

That brings me back to the reasons why people do not claim pension credit, some of which are within our gift. We need to inform people about it, and we are working with Age Scotland on what we might do in that respect. Next month, the BBC will send 450,000 pensioners in Scotland a letter, telling them that, unless they receive pension credit, their free television licence will come to an end. I am looking forward to that, as we expect to get a lot of calls at that point from people seeking information about their pension credit entitlement.

I have one further thing to say. The pilot that I mentioned was held in 2010, and, because we had to aggregate the information, its accuracy was not terribly high. We have been working closely with HMRC, particularly on data feeds and things like how much occupational pension someone is actually receiving, because, for the pension credit caseload, that is a much bigger issue than earnings, for which the automation is much more established.

I am tempted to ask a lot more questions about that 2010 pilot but I will not, because of time constraints, although I think that the committee would like some more information on that. Was it a Scottish pilot?

No, it was a Great Britain pilot. I can provide the committee with the published evaluation.

What was the cohort? How many households were involved?

Two thousand.

We would be keen to see that.

I will stick with pension credit for now. I will not make an issue of this, but the committee has made an issue of it in the recent past and I want to put it on the record, because my constituents would disappointed if I did not. The committee was pretty upset at the UK Government’s decision to withdraw pension credit from mixed-age couples, meaning that, if the statistics are accurate, some households will lose up to £7,000 per year. There will be 5,600 of those households in Scotland by 2024—I will leave that sitting there. If I had not said that, people would ask me why I did not say it. The committee thought that it was just wrong.

Another unintended consequence of universal credit and pension credit is that, by 2023, a pensioner who is receiving housing benefit might have to claim it through pension credit as housing benefit is phased out and becomes the housing element of universal credit rather than a stand-alone benefit. Is that another opportunity to do some more work on who is entitled to pension credit in the first place? Are you concerned that it might lead to another unintended consequence, which is that some people who were getting support for their housing costs might not get it after that?

I have given a lot of thought to that issue. What was said in the earlier evidence session is important in this respect.

In the UK, 16 per cent of pensioners are in poverty. If all those pensioners claimed pension credit, housing benefit and the council tax reduction—especially the council tax reduction—that would reduce the 16 per cent to almost zero. The relative take-up rates for those three benefits are as follows: about 80 per cent of pensioners who are entitled to housing benefit claim it, about 60 per cent claim pension credit and only about 50 per cent claim their council tax reduction. If housing benefit and pension credit are brought together, there will be the opportunity to increase the take-up of a unified benefit.

We have done quite a lot of research into why there is that differential, and we think that one of the reasons is a matter of perception among relatively low-income pensioner households. The local authority pays their housing costs and the payment generally goes straight to the landlord, so the pensioner sees the good rather than the cash. With pension credit, however, they have to phone the pension centre at the Department for Work and Pensions and answer what some people consider to be intrusive questions about how much money they have in their bank account. That kind of thing puts people off, along with the more general stigma around means-tested benefits, particularly for much older pensioners.

As you said, convener, there will be that opportunity but, for clarity, it will not happen before the end of 2023.

Is that the earliest that it would happen?

Yes.

That is very helpful. Let us stay positive. If there is an opportunity there, we want to help and push some of that forward as a committee, because that is our core business. You will understand why there is a nervousness among certain groups. The last time that pension credit was reformed, it was a negative reform that was detrimental to many households, who suffered financially, so people might think that this next reform could also be detrimental.

If there is an opportunity for us to work together more closely on aligning entitlements and maximising uptake among pensioners in poverty, the committee is all for that—of course we are.

You will have heard, in the previous session, some thoughts about how benefit take-up campaigns could be improved and made more targeted. You will also have heard that there was a call on the DWP to share data more routinely with councils. A bit of work has been done on that, particularly around fraud, and councils appear to be keen to do more in that space. Is the DWP open to doing more joint working with local authorities to use data to target people and ensure that take-up is maximised?

Thank you for allowing us to sit through the earlier session. The discussion on data and automation was really helpful and thorough. I was struck by something that came up both in what the witnesses said and in the questions that committee members were asking, which is that it is not necessarily technology that is the barrier; the barriers are more about policy or legislation. Sometimes, the barriers are more perceived than real: we heard about GDPR being seen as a block, whether or not it actually is a block. We also heard a witness speak about people having reasonable expectations about what happens to their data, and about transparency—the need to be clear with people about what happens to their data. I was quite struck by that.

I know that you will have the cabinet secretary before you next week. My understanding is that the Scottish Government has done some work with local authorities to see whether they can identify, through the Digital Economy Act 2017, an appropriate legal gateway that will allow data to be shared between the DWP and Scottish local authorities. That process has not yet concluded, but that may well provide an avenue for the sort of joint working that you are talking about.

Obviously, universal credit will be critical in this area. You heard earlier that there is a mine of information in universal credit claims that would enable access to other benefits. We will certainly pursue that.

Another issue that we have looked at is the stigma around universal credit generally. I will not rehearse the reasons for that stigma, but we can say, without a doubt, that it exists. How can you ensure that people are not discouraged from claiming universal credit? The perception—and sometimes the reality—is that it can make people worse off and that claiming it can be difficult and lead to hardship. What action are you taking to address that stigma?

I will start, and Derek Kilday might want to come in.

You raise a really important issue on which we have done quite a lot of work. We have tried to put out into the system clear information about what universal credit is and what people might be entitled to. We have worked closely with partners: one really encouraging partnership has been with Citizens Advice Scotland, and we have had partnerships with local authorities, schools and housing associations. That is a really important step.

A couple of weeks ago, I went to Shettleston—on Derek Kilday’s patch—and work coaches there talked to me endlessly about how they wanted the jobcentre to be in the community and how they were going to all sorts of organisations to talk about what the DWP can offer them, instead of its being just a door that people have to enter.

10:30  

A third part of what we have done involves the reassurance and the support that is given by work coaches in the jobcentres. Those are the three levels: information giving, at the top end; working with other partners, which is important for us; and the local, face-to-face reassurance that we give to our customers.

There are two elements to the issue. We do a lot of direct work, but, in terms of take-up and support, we also do a lot of indirect work. That involves working with partners and what Katie Farrington alluded to about putting work coaches into the heart of the community, where they can reach these vulnerable groups. There are examples of that all over Scotland, but I will give you an example from Glasgow.

We work with Glasgow City Council on vulnerable persons relocation. We and the social work department try to provide a smooth transition on to benefits, and we liaise with the council to make sure that the individual has completed their claimant commitment and the other necessary parts of the application process. In that way, we provide a tailored support to the individuals, which has been successful. If any problems arise, we are able to deal with them. There are drop-in centres in Partick and Shettleston, and we work with Glasgow City Mission and others. In the past year, we have been active with the Simon Community to prevent homelessness, and we have run upskilling sessions with it and its partners on that subject.

The rurality of Scotland was touched on earlier. In rural communities, we adopt a multiagency approach with our key stakeholders, working in true partnership with the local authorities. Oban is one of my patches, too. From there, we provide outreach into Rothesay and that whole community, working in places such as libraries and local authority buildings. A remote claimant multiagency approach is also being developed in Wick, and we plan to move that out towards the Western Isles. We regularly send work coaches on ferries to these places, so there is a drop-in presence.

That is helpful. For the record, can you confirm that, on the sharing-data issue—you pointed to what you described as a legal gateway—you are saying that the DWP has no policy barrier to sharing data with local authorities and that any issues in that regard concern the mechanics of doing that? Is that accurate?

The DWP wants to ensure that it is sharing and holding its data legally and in line with the proper requirements. I do not perceive that we have a policy objection to sharing data, but we have a legal duty to hold data safely and securely and to share it appropriately, in line with legal gateways.

There are a few bids for questions. It is only fair to let Mr Balfour in next, as he was last in the previous session.

Before I ask my question, I would like to clarify what was just said to Shona Robison. I understand the issue with data protection, but I am slightly concerned that it is used not only by the DWP but by other Government bodies, such as the NHS, as an excuse to not share information. Are you being overcautious in that regard? If so, why?

I do not perceive that we are being overcautious. As you heard from the previous panel, everyone wants to ensure that they are compliant with the new legislation and that they are doing all the right things.

There is something important about reasonable expectations. If my GP refers me to a hospital, I have a reasonable expectation that my data will be shared, and I would most likely be quite frustrated if, when I turned up at the hospital, the people there did not have access to my records and I had to repeat my story over and over again. However, if I give my data to the hospital, do I have a reasonable expectation that it will be shared with other parties? I think that that is the point that members of the first panel were talking about. I think that the deputy convener made the suggestion that, if we can ask people whether they are happy for their data to be used to tell them about other benefits to which they might be entitled and they tick that box and give consent, anyone who sought to use that data for that purpose would be on a much more secure footing than they would be if we arbitrarily shared it.

That leads on to my main question, which is about your process of checking the information that you have on people.

As I have said many times, I am a client of the DWP—I receive PIP, which comes into my bank account every month. Do you ever look at my account and say, “That’s Mr Balfour; is he entitled to any other benefits?” I am talking about the benefits that the DWP is responsible for, not the devolved stuff. Would it be possible for you to do such an exercise? That would involve checking whether someone who received one benefit was entitled to any other benefit, rather than their having to ask, “Am I entitled to that?”

That is a fair question. I am not aware that we do that, but I will turn to my colleagues in case I am incorrect.

I will give an example that relates to PIP and attendance allowance. We cross-check to see whether the person would get an increase in their pension credit as a result of their receiving a disability benefit or having applied for carers allowance, in which case they would be entitled to a carers premium. We do that check within the system.

However, there are complications. The severe disability premium was mentioned earlier. A couple can choose whether they want the partner with the disability to receive the severe disability premium or they want the other partner to receive carers allowance; those two benefits are about the same amount of money. The couple must have a discussion about how they want to do that. Therefore, there are some things that we cannot automate, but we can get in touch with people to tell them that they can think about that.

I suppose that my point is that no one has ever been in touch with me to ask, “Have you thought about applying for carers allowance?” I appreciate that responsibility for that has been devolved, but at no point in the past 10 years has anyone from the DWP written to tell me that I might be entitled to carers allowance if I met the criteria and to ask whether I had thought about applying for it. Should you be a bit more proactive about approaching individuals?

Many people might be on PIP or attendance allowance, for example, but they might not know that they could also be eligible for carers allowance. When you send people the annual statement that says what they are entitled to and how much they will get, could you not include a booklet setting out the other benefits that they might be entitled to? I appreciate that that would be hard to do from an administrative point of view, but it could lead to greater uptake.

Perhaps we could have another pilot similar to the one that Mr Latto mentioned in relation to pension credit. That was a good question, Mr Balfour.

That is twice in one day that you have said that, convener.

As the previous witnesses said, at this stage, operationally, we rely on moments in people’s lives. Mr Balfour is absolutely correct. We use those moments in people’s lives as our trigger points. As we move to widen the expanse of people on the UC system, those life moments will become more frequent. Mr Balfour asks a very good question, which we will certainly take away and consider.

Does anyone have any additional comments on that, before I bring in other members of the committee?

Only to say—to remove my pension hat and put my devolution hat on—that there will be an opportunity to do that, because the disability benefits and the carers benefits will be replaced in Scotland.

Absolutely.

We can put that question to the cabinet secretary next week.

Kate Farrington saw where I was going, and I thank her for her reply.

Can we make such data sharing happen? How do we make it happen? It seems to me that it might solve some of the problems that people are arguing about. It is our experience that, not just in public agencies but public institutions, there has been a complete closing down of the sharing of information at the most basic level because of confusion or fear over what might happen if data were shared.

Can we make that process begin when people are asked to share data? Can you do that?

I am pausing because I am thinking about your question, which is a very fair one. The point that sits behind it—that there is a fear of data sharing—is also fair. Could data sharing happen? I am sure that that would be feasible. How long would that take? How complicated would it be? Sitting before you today, I cannot give you promises on those questions.

Given that lots of benefits are now being devolved to Scotland, there is an interesting question for the committee—I assume that this is behind some of the questions today—about the right action for the Scottish Government to take, within its powers. That might be quite separate from what the DWP does or does not do.

I can give an example from some time ago relating to the purposes of sharing data. Our work on the warm home discount scheme, which I know is coming to Scotland, involves data matching between pension credit and the energy suppliers’ customer records, so that they can generate a £140 discount each year on people’s electricity bills. We took primary legislation to do that; we do it without the consent of the pensioner. Very occasionally, we get complaints about that, but normally we get letters that say, “Thank you very much.”

Conversely, some time ago, we looked at a pilot with Cheshire Fire and Rescue Service, which was quite interested in knowing the names and addresses of vulnerable elderly people, so that it could check that they had smoke alarms, fire systems and so on. That was more difficult, because some people do not want such information to be shared. We found an alternative that involved identifying clusters of roads where there happened to be large numbers of people with those characteristics, but we did not share customer data. In that case, we did not even think about asking individuals whether they wanted a fire brigade officer to come round to check their smoke alarm for them, because that was considered to be too sensitive.

Automation between benefits systems is one thing but, as we move out, there are gradations according to the sensitivity of the data.

We are here to discuss a range of issues but, at the heart of them all, is the fact that we want to ensure that everyone receives what they are entitled to receive. Clearly, that is not happening to the extent that we would all wish. I want to ask specifically about the underpayment of disability benefits. We are aware that £330 million-worth of PIP was unclaimed due to claimant error, which seems massively frustrating and very sad indeed. Why was that the case? What can the DWP do to help to prevent that?

Of course, I should start by saying that such benefits are being devolved to the Scottish Government so, in future, those matters will be for the Cabinet Secretary for Social Security and Older People and her officials.

As people have said previously, and as we have reiterated, it is very difficult to calculate the non-take-up rate of a benefit when a discretionary exercise is done to determine whether people meet the conditions for it. In the case of PIP, the condition is that people have care or mobility needs.

I confess that none of us on the panel is an expert in PIP, but I imagine that the figure that has been given is an estimate relating to people who could have got the higher rate; they would have got some money through PIP, but they could have got a higher rate if they had told the DWP that there had been a change in their circumstances. It is quite difficult for the DWP to know when there has been a change in somebody’s circumstances unless they present that information to us.

There are some things that we could do—I am thinking aloud now. If we receive a claim for carers allowance from somebody who receives a personal independence payment, that might be a trigger for us to think that the circumstances of the person with the care needs have changed. I am not sure that that happens, but that is the type of thing that we were thinking about earlier.

However, in the case of benefits for which we are reliant on the claimant telling us that there has been a change in circumstances, the best that we can do is inform people, as clearly as possible, about the circumstances in which it might be in their best interests to contact the DWP and say that there has been a change in their circumstances.

This relates to the media, too, but some people feel that, if they contact the DWP, there is a 50:50 chance that they might get less, instead of getting more. The more clarity we can give to people about the circumstances in which they can contact us, the better.

10:45  

Yes—that point is well made. You will be aware that there are people for whom the system is complex and engaging with it can be difficult. I remember that, when the deputy convener, Pauline McNeill, and I attended a session with universal credit claimants in Musselburgh a year or two ago, a young man who had health issues was receiving messages on his mobile phone about sanctions while we were sitting with him. Given the complexity of the engagement, there are some obvious reasons why people might underclaim. I am keen that we do as much as is needed to make the system simple.

The benefit take-up strategy that the Scottish Government has published with respect to some of the payments that come under the Social Security (Scotland) Act 2018 outlines how it intends to promote take-up of devolved social security. I would like to understand what the DWP’s equivalent strategy looks like.

We looked with interest at the strategy that you mention, and one thing that struck me—perhaps there should be no surprise about this—was the consistency between some of the points that it raises and some of the things that are important to us. It says quite a lot about starting with the user, building around the customer, removing barriers and giving clear information. Those are all things that we whole-heartedly support and we are seeking to build them into the benefits that we are designing.

We do not have an equivalent written strategy, but the thing that struck me when I read the strategy was the number of common threads that exist—things that the Scottish Government is seeking to achieve that we also support.

I am somewhat surprised to learn that you do not have a written strategy. Would you like to change that?

Let me start with universal credit, because it has been designed implicitly to be simple and built around the customer. There may be many different opinions about how successful it has been and we are seeking to continue to improve it and to learn from the experience as it is rolled out, but it was designed to be simple, to be built around the user and to break down some of the barriers that existed in the legacy benefits system. That system contained many different, complex benefits and it had thresholds built in whereby, if somebody worked for more than a certain number of hours, the support that was available to them fell off a cliff. When I looked at the strategy, it seemed to me to be extremely coherent with some of the design principles that sit behind UC, which I know have been well rehearsed with the committee.

I think that my takeaway is that, given the vast amount of unclaimed benefit, I am surprised that the DWP does not have an equivalent written strategy, but perhaps we can explore that further later.

The committee will certainly want to return to that. The Scottish Government strategy should not sit in isolation. Given that we have various tiers of government, from local authorities right through to the DWP, with Social Security Scotland in the middle, they should dovetail in their benefit uptake strategies. There is a little bit of concern that the DWP has not developed a written strategy, although I appreciate that it works under instruction from the UK Government. The committee will have to consider that.

This might be an appropriate moment for me to say that we work closely with our Scottish Government colleagues and have a very good relationship with them. I know that Andrew Latto meets colleagues regularly, and Derek Kilday also talks to them. There may well be differences of view about various things, which is entirely appropriate, but we do have a mechanism to discuss those and resolve them where we can.

I apologise for repeating what I said earlier, but we visited Social Security Scotland yesterday, and it was clear about its high-quality day-to-day operational relationship with the DWP. I put that on the record again.

We are talking about how to increase benefit uptake, and universal credit has been touched on. It is fair to say that universal credit was a mammoth change in the welfare landscape across the United Kingdom. One of its main policy aims was to simplify the six legacy benefits and increase uptake. How has going from the six legacy benefits to universal credit changed uptake? What improvements in uptake has universal credit brought to people?

I will start, and Derek Kilday may want to add to what I say.

The big benefit of universal credit is the one that you have just referred to—it has brought together the six legacy benefits into one benefit. Historically, some benefits to which people were entitled were claimed and others were not. The idea with universal credit was to bring all the entitlements together in a simple fashion. I ask Derek Kilday to comment on this, but I believe that the experience of that has been pretty good on the ground. Generally, people find that the approach is simple, that they get their payments pretty accurately and that they take up what they are entitled to.

We have worked long and hard on that since our previous committee appearance, when we mentioned the good progress that we had made on paying people correctly in full and on time. For the past few months, we have consistently hit that for 95 per cent of people who claim universal credit. Across the board, 95 per cent are paid in full and on time. We are justifiably proud of the big improvement that we have made.

In light of Ms Johnstone’s earlier comment, it is worth putting on the record that we are in a different place with sanctions from where we were two years ago. We do not apply sanctions indiscriminately; we have a fair and effective look at things. In August 2019, only 2 per cent of people on conditionality were sanctioned. That is a very small number. The National Audit Office’s “Benefit sanctions” report stated that 53 per cent of ESA claimants who were looking for work said that sanctions are more likely to make them comply with the conditions. As members can see, we do not apply sanctions indiscriminately—indeed, we never did—and we are now tracking that. After the previous committee meeting, we put in place a management check mechanism for all of the UK and specifically Scotland. A check is done every day to ensure that we do not indiscriminately apply sanctions.

I am sorry if my question was not clear, but I specifically asked what improvements in uptake you have seen as a result of the move from the legacy benefits to universal credit.

That is difficult to say. We know the numbers, but it is difficult for us to track the data pre and post the changes.

One of the big policy objectives of universal credit was to increase uptake. It is surprising that one of the big policy objectives and the means by which universal credit was promoted and sold to the public has not been tracked. Do you have any indication, even from the East Lothian or Highlands pilot areas, of the uptake of universal credit compared with the uptakes of the legacy benefits?

I do not have that information with me, but we will happily take that question away and write back to the committee about it.

Obviously, universal credit has been a huge change in the landscape of social security assistance, and it is a good example for the Scottish Government and other agencies to look at to see whether it has been beneficial and has increased uptake. Therefore, I urge you to look at those figures and, if possible, provide them to the committee so that we can include them in our consideration.

I know that the statisticians in the department who generally provide our figures on benefit uptake are looking hard at universal credit to ensure that we accurately measure uptake and publish those statistics, as we publish other statistics. That was in the paper that the DWP submitted to the committee.

At the moment, we have two challenges on universal credit. One is that until recently it was still being rolled out, so the landscape has been quite fluid and mobile as the new system was rolled out across the country. Secondly, because universal credit includes an element for disabled people, we need to know who would be entitled to the disability benefit in order to know whether the people who are entitled actually take it up, which, I appreciate may sound like a complex technical issue but it is one that we are working on. As soon as we have robust data and evidence on that, we will publish it.

That would be helpful, not just in relation to universal credit but in relation to the entitlements and benefits that are being devolved to the Scottish Government and Scottish Parliament, because low levels of uptake would have an impact on the block grant transfer from the UK Government to the Scottish Government. If the UK Government performs poorly on uptake of reserved benefits that are then devolved, any additional uptake at that point would, understandably, have to be funded from Scottish Government resources. It is important for the committee to get more detailed information on that, so that would be helpful.

We will move on to the next line of questioning.

When we questioned the previous panel about what could be done to provide a more seamless experience for people claiming benefits, we got the interesting advice that the Scottish Government should focus its attention on local authorities because, to paraphrase, we might wait a long time for the DWP and the UK Government to provide some of the change that is needed. Do you have a response to that?

It is not for me to tell the Scottish Government what to focus on, but the benefits that are being devolved are an opportunity and it will be interesting to see how they are developed. I do not necessarily recognise the caricature of the DWP being closed and unhelpful, but I recognise that data sharing can be a barrier and it can be simpler and easier within one organisation rather than across boundaries.

Does anyone want to add to that?

On the data-sharing agreements, as Katie Farrington mentioned, the legal gateway would be the best opportunity to get a successful outcome.

The concern that we all share is to ensure that people do not fall through the cracks between systems. Are there concrete examples of things that could be improved or developed? I am thinking particularly of people being referred by the DWP on to Scottish benefits. Is there anything that you can see changing or being developed there to avoid problems?

That is more about the operational forum that we mentioned earlier, at which the most senior people meet regularly. There is a Scottish meeting between Social Security Scotland and the DWP, which is an opportunity for us to understand and share that type of information. There have been strong relationships thus far, and I see that strong partnership working continuing while we put things on the table.

Obviously, you will take a different view, but that was not the tenor of the evidence that we received in the previous session. I am curious to know whether you have any comment on the concern that was expressed in the previous session about the lack of data sharing and referring on. Do you recognise that criticism?

Where there is a legal protocol to do that, we most certainly do it. Where we can, we will work with the Scottish Government. It has been recognised that we have passed over things and shared the data, so I do not recognise what you have said as the complete picture.

Such matters are complex and, as everyone on the committee knows, big IT projects are complex and can take a long time. There is no general single one or two-year solution for such things. I was interested in the point that, if we are going to have such a system, we have to think about doing it in phases and building towards it.

11:00  

I will mention our experience in relation to the Scottish Government benefits. Those benefits are the carers allowance supplement, which is automated and which is an interim measure until carers assistance comes in, and the funeral support payment and the best start grant, which derive from a universal credit entitlement. As far as I know, that is functioning well. There will be the job start payment, which is a completely new Scottish payment. We are working with our work coaches on how to deal with and use that potentially useful tool in Scotland.

We are now gearing up for the replacement for the disability living allowance for children later in the year. One aspect of the Governments’ joint commitment to have a safe and secure transition is that the DWP will not turn off anything until the Scottish Government is in a position to turn things on. That includes the derived entitlements through which the parents of a child may be entitled to carers allowance. At the moment, we are delivering that on behalf of the Scottish ministers. There could be premiums in means-tested DWP benefits, too.

We are working closely with the Scottish Government at all levels—legislative, data sharing and operational—to ensure that people in Scotland receive the service that they need to receive.

Mr Kilday, you mentioned big IT projects. Do you accept that it is time to build into the concept of big IT projects the fact that people want systems that take the issue into account? I understand very little about IT projects but, as a layperson, I know that we need to build something into future IT projects that makes it easier for people to get their benefits. Will you put it on record that you accept that?

That is a fair comment. We aspire to do that. Like the Scottish Government, we rightly try to put citizens at the heart of everything that we do.

Obviously, this session is about benefit take-up. I am struck by the irony that the example given at the start of this and the previous session was how to increase take-up through the elimination of another entitlement—TV licences for the over-75s. However, we have heard, I think in response to Alison Johnstone’s question, that there is no written strategy for increasing benefit take-up. We also heard, in response to Mark Griffin’s question, that there is no way of tracking whether take-up has increased compared to take-up of the benefits that preceded universal credit. Alison Johnstone was surprised about that; I am appalled by it. The third leg of what we have heard is about not knowing—and the difficulty in knowing—the level of non-take-up and the reasons for that.

I am interested in what, if any, research the UK Government or the DWP has done to understand the position. I understand that, for various reasons, it might be difficult for systems to tell you what the lack of take-up is, especially for discretionary benefits, but surely the way to plug that gap in understanding would be through research.

You picked up on my response to Mark Griffin. We have not yet published the data on universal credit take-up, but we absolutely intend to do that as soon as we can do it in a way that is statistically robust, and my colleagues are working on that.

I have some research, although it is not precisely on the topic that you mention, but it is interesting and I would be happy to share it with you. It is data from the universal credit full service survey that was published in February 2019, which looked at the reasons why people delayed their claim for universal credit. That is not quite the same as not taking up universal credit, but there are useful parallels.

In answer to the question “Why did you delay making your claim?”, 28 per cent said, “I did not know what to do”, 27 per cent said, “I thought I’d get another job quickly”, 21 per cent said, “I thought I wouldn’t qualify”, 16 per cent said, “I have a medical condition that prevents me from applying”, 12 per cent said, “I thought my old benefits would start again”, and 6 per cent said, “Lack of support”. There are some interesting trends about whether people understand what they are entitled to, which I know the committee has been interested in, whether they understand how the system works and the fact that sometimes people put off making a claim because they would rather see whether they can get back into work before doing that.

There are complex issues, and I know that the committee has looked into them in some detail in your inquiry. I am happy to share that evidence, if that would be helpful input.

That would be useful, because the delay is part of the lack of take-up. We want to get a better understanding. In previous sessions, we discussed cultural barriers—people feeling that they were not really entitled to benefits—and the barrier of people feeling that the system is impossible to navigate. Given the resources that are spent on social security, it is fundamental that the DWP and the UK Government should have an understanding of why it is not achieving the aims that it is trying to achieve. Researching the level of non-take-up and the reasons for it is fundamental. In the previous session, we were told that no research has been carried out in the past 10 years, although the example that you gave might be an exception to that.

In the previous session, we heard that an obligation was written into the Scottish legislation that requires the Scottish ministers to maximise benefit take-up—I am paraphrasing, so that might not be exactly what it says. Does that obligation have a counterpart in UK legislation?

There certainly is not a direct counterpart. There is a general principle that we need to inform people of their entitlements, but there is nothing that says that the minister must report to Parliament regularly on whether we have achieved 100 per cent of take-up of benefit X. That does not exist in UK legislation. Part of the reason is to do with how we would demonstrate that. Presumably, the Scottish Government will find a way of demonstrating it.

Research might be a good idea.

Indeed.

Can I say one other thing on research? Another area that I am responsible for is attendance allowance. In the context of social care reform in England, we are looking not solely at the benefits system, but at how people cater for their care needs and remain independent in their own homes, and the extent to which they are assisted to do that by the social care system and the benefits system, through attendance allowance, for example. We are also looking at how those benefits interact with free personal care in Scotland. Obviously, from 1 April, that will become a matter for the Scottish ministers, but I thought that I should mention that we are doing research in that area as well.

I would like to follow up on that, after which I will bring the deputy convener back in.

Mr Latto, although you said that there is no statutory duty on the UK Government and the DWP to have a strategy to maximise the uptake of benefits where possible, you said that there is a “need to inform”. Can you explain what you meant by that?

There is a general obligation on public bodies to inform people of their entitlements. I should say that we report through the annual accounts, and how we administer the benefits system is audited by the National Audit Office, so the UK Parliament’s Public Accounts Committee has a means of holding the DWP’s accounting officer and ministers to account for the way in which we deliver the services that we are obliged by law to deliver.

Is there data somewhere on what additional income has accrued to individual claimants in circumstances in which, when they claimed for benefit A, the need or duty to inform kicked in, and they came out with benefits B and C? I am sorry to be so simplistic about it, but how can we monitor whether that need or duty to inform is doing what it is supposed to do?

My sense is that the best answer to that will come through the statistical work that I mentioned, perhaps together with the sort of research on delays that I have just talked about. I suspect that the truth is that, day in and day out, our work coaches ask people sitting in front of them, “Oh, have you thought that you might also be entitled to another benefit?”, but that we are just not collecting information on that systematically.

That is helpful, because my next point was going to be about that.

As I am about to mention yesterday’s visit again, I should clarify that we did not go to Social Security Scotland; we went to the Scottish Government delivery team, who work with Social Security Scotland and who are based at Victoria Quay—we met the key delivery leaders there. During that conversation we heard, as we have heard previously, about the good working relationships between the DWP and the Scottish Government and Social Security Scotland. It was clear that there is an understanding that Jobcentre Plus work coaches will, we hope, be willing to not just inform claimants of reserved benefits but signpost them to entitlements to devolved benefits. That is positive, and we will certainly ensure that Social Security Scotland does likewise in relation to entitlement to reserved benefits when people engage with it. We hope that the system will work in that seamless way.

What training are work coaches getting or do you intend to give them to ensure that, in what is a dynamic system that is entering a period of dramatic change, they are up to speed and can give the appropriate information? The Public and Commercial Services Union has put on record at the committee significant concerns about strains on staff in jobcentres. Therefore, as well as answering my question about ensuring that staff are informed enough to convey quality information to clients, will you say whether they have time to do that in an empathetic and effective way?

As the digitised UC case load grows and grows, that creates more time for front-line staff in jobcentres. During the past year, I conducted an individual learning needs analysis across every one of our employees in Scotland, and each person will have a personal learning and development plan. Part of that is about understanding signposting for benefits. Every week on a Wednesday, we take a significant time out of our business to upskill our people on the front line in relation to that sort of information. I am confident that we have the capacity and the ability to get people signposting in the right manner.

Will that be audited and will information be collected systematically? We might want to ask the PCS what it thinks, and it could say that, if the system is getting more complex, that places an additional demand on work coaches’ time. In my experience, they are keen to convey that information, but they can do only so much with the 10 or 20-minute slot that they have with clients. How will the situation be monitored, and how could the committee scrutinise some of that monitoring?

It is a difficult one to monitor. It would be costly to put something like that in place, which is why I have introduced the individual learning needs analysis. I will rerun that process and, through that, front-line coaches will be able to tell us whether they are having difficulty understanding the new devolved benefits and how they interact with other benefits. That is the tool that we have in place in each of the areas in Scotland to allow my service leaders to understand their training needs across those areas. We will continue to monitor via that route.

Katie Farrington gave a number of reasons why people delay applying for universal credit, but she did not mention that one reason is that they might lose their child tax credit payments. A minority of people might do better, but most people fear that they will do worse. I have previously raised a case involving someone who had been in work for 30 years and who was changing jobs. They were within the four-week roll-on period that is allowed in that situation, but the work coach did not advise them of that and, as a result, the person lost out. I put it on record that I am still in dispute about that.

That connects to the point that Derek Kilday made. Obviously, the system is still new—that is understood—but the DWP seems to be absolutely rigid about this. You should be aware that people are calling it the lobster pot. Their experience, and that of many MSPs and their constituents, is that if someone makes the wrong decision—such as by signing up for universal credit, only to find out that they are not entitled to anything—they are not allowed to go back. That happened to the claimant in the case that I mentioned, even though she was within the four-week roll-on period, and I will continue to fight with the DWP about that.

It concerns me that you do not seem to know that that is one of the reasons for people delaying applying for universal credit. If someone is likely to lose out on their child tax credits—which for some people are pretty generous—why would they come forward? Are you aware of that issue?

11:15  

That has not come through in the research that I mentioned and which, as I said, I am very happy to share in writing if it would be helpful to the committee. I am aware that there is a human worry that if people move across to UC, they might be worse off or something might change in those circumstances. I also know that, where they possibly can, the work coaches are putting in a lot of time to reassure people about the benefits that universal credit can bring them. You have raised an important point, but I do not know the detail of that case.

Have you heard the phrase “the lobster pot” before?

No.

Right. The system is like a lobster pot, because—

What you mean is that once someone is in it, they are trapped.

I am sorry—the phrase is not mine. It is used by a lot of people, although it is perhaps not the best description. People feel that, once they are in the system—which they did not want to be in in the first place—they cannot get out of it. In my opinion, that is deliberate policy. Perhaps there is nothing that you can do about it, but I think that it is deliberate, because people will get less benefit as a result. I just wanted to make you aware of that.

We will always have isolated cases, and we will certainly pick up on the case that you mentioned and have a look at it, but our experience has been that many people are better off as they come on to universal credit.

Sure—some people are.

We also spend a lot of time explaining its benefits. However, there is an onus on politicians, the media and everyone else in Scotland to take away the toxicity of the universal credit brand. We have some very good success stories about people who have been helped and supported through claiming it. Universal credit is an in-work benefit. We have a duty to create work and wealth in Scotland, and universal credit is part of that process.

It is incumbent on us all to look at isolated cases for what they are. However, we will be very happy to pick up on the details of the case that you mentioned.

Thank you.

That is helpful. The committee is always keen to show that there are two sides to every story and to highlight good practice where it exists. However, we absolutely need to scrutinise cases in which there are concerns, and the media will always pick up on the bits that they tend to pick up on.

Before we close our evidence session, I would like to check a few points. The committee will have to publish its report on its inquiry relatively soon. Much of the discussion in our evidence sessions has referred to automation. Can you give examples of situations in which the DWP has actively pursued automation or is considering it for the future? Are there some situations in which there is an opportunity for better use of passporting? I would like to get evidence on that on the record. Can you give examples of where a proactive approach has been taken?

The DWP completely embraces new technology. We were looking at the UC build programme from the point of view of automation, as well as other aspects such as customer service, reducing costs and increasing efficiency. A lot of work has been done on reducing fraud and error through such automation. We have a number of strands of work to explore the options that might be available to us to automate services such as telephony, to reroute information or do things differently. We have already rerouted some telephony in Scotland, which made a magnificent improvement, because it meant that people’s calls were not ringing out. It also improved how quickly we could answer our phones, and our average answering speed is now down to a record time. We are looking at whether the technology is available to allow our staff to interact with people in the more complex, difficult cases.

I am not sure that that is quite what I meant by automation. You have described the use of technology to make the behind-the-scenes aspects at the DWP work more efficiently to improve service delivery. I see what you are saying, but have you had any policy thoughts about how to use IT and automation to maximise benefit uptake rather than to deliver the system more efficiently?

I can give two examples. Earlier, I mentioned the warm home discount scheme, on which we work with energy companies to generate a passported entitlement. The other is cold weather payments, which we automate between ourselves and the Met Office for cases in which people are on relevant benefits and the trigger has been reached at the relevant weather station. We are working with the Scottish Government on its replacement for the current such scheme.

Thank you. Are there any other examples that you might want to put on the record? Earlier, Mr Latto made a helpful reference to the pension credit pilot in 2010. I suppose that we could frame that as a benefits uptake initiative, because it clearly was: it produced a 9 per cent recurring benefit. A decision could be made about whether that represented value for money. Have you other examples of successful benefits uptake campaigns that have existed since 2010?

We have not yet talked about it, but perhaps we should have mentioned the very current example of the help to claim scheme, which I am sure that many members will have seen in action. It is being delivered through citizens advice bureaux—here, that is being done through Citizens Advice Scotland—to help people to claim money to which they are entitled. We have also done a lot of work on information giving. I do not know whether any of my colleagues wants to add anything on that.

We are running short of time, so perhaps you could write to us about help to claim. The committee has looked at that in some detail, but we still have concerns about operational protected dates of claim for claimants for whom local authorities were able quickly to take basic core details, after which any further details that came later were backdated to day 1. We have heard anecdotally that such backdating has not always happened. I know that Citizens Advice Scotland is very keen to ensure that that happens, and is doing its best on that.

The committee is also keen to receive data on the success or otherwise of the help to claim scheme. I understand that the relevant data is being fed into the DWP system for analysis. We are keen to see that as quickly as possible. We have heard significant concerns about the strain that is put on services by the six-week delay until the first universal credit payment is received. Financial support for help to claim is provided through the citizens advice network, but vulnerable people continue to require assistance way beyond that. We need to see mapping work on what happens after week 6—we cannot just walk away from people.

The committee will be happy that you have mentioned the example of the help to claim scheme, but we have enduring concerns to which we would like to return. If you would like to make some brief comments on the matter, please do, but I am quite keen for you to write to us on that.

My understanding is as you have described, convener. The first payment date has been an issue, but people have been working together and we think that we are now in a much better place.

I should say that, as well as having the help to claim scheme, we continue to offer support for claimants through our jobcentres. For example, we have a freephone telephone line that is available long beyond the first six weeks of a claim.

Because of time constraints, I had better not go on to talk about communication with parliamentarians and jobcentres. The committee is interested in the on-going work on that, which we will undoubtedly discuss in due course.

Time is upon us. Do members have any further questions?

It is not really a question, convener; it is more a request for information. An interesting point was made about using the Digital Economy Act 2017 to facilitate data sharing, which is at the heart of this issue.

If it is the case that the DWP, the Scottish Government and local authorities are looking for a pilot area, perhaps a very small local authority area in the Central Scotland region might be appropriate. [Laughter.] I extend an open invitation to our witnesses to come to Clackmannanshire at any time should they want to look into that further. In any case, information on where any such pilot might be conducted would be useful.

Our witnesses should definitely do that, but only after they have been to Maryhill and Springburn in the Glasgow local authority area.

We are keen to have on-going dialogue and a constructive relationship with the DWP. We will not always agree with it but, in our scrutiny, we should be respectful. We would like to see our witnesses as repeat visitors to the committee’s meetings so that we can keep such dialogue going. I thank them very much for taking the time to attend today.

That concludes agenda item 2. We will shortly move on to agenda item 3, which we have previously agreed to consider in private. I therefore move the meeting into private session.

11:24 Meeting continued in private until 11:33.