Social Justice and Social Security Committee

Meeting date: Thursday, December 22, 2022

Official Report 644KB pdf

---

Social Security Scotland

The committee will now take evidence from Social Security Scotland on its performance and operation. I welcome to the meeting David Wallace, the chief executive; James Wallace, the deputy director for finance and corporate services; and Janet Richardson, the deputy director for client services delivery. I invite David Wallace to make any opening remarks before we move on to questioning.

Good morning. I am grateful to the committee for inviting me to give evidence today. With me are Janet Richardson and James Wallace, whom I believe most members have met.

It is just over a year since I last appeared before the committee, and once again it has been a hugely significant year for Social Security Scotland. We have launched the adult disability payment, which is the most complex benefit that Social Security Scotland has yet delivered. I was pleased to welcome some committee members to our Dundee headquarters in the spring to hear about our preparations for that. By the end of October, we had received over 18,000 part 2 applications, and we are working through them as quickly as possible.

That has come alongside work to transfer existing awards for PIP and child disability living allowance to Social Security Scotland. We have already transferred over 34,000 clients, making sure that they experience no breaks in payment. Last month, we extended the Scottish child payment to children under 16, and we have received 123,000 applications in less than a month.

Social Security Scotland has grown rapidly over the past year. With over 4,000 staff now in post, we have completed our volume recruitment, which means that we are well positioned for our live benefits and the new winter heating payment from next February. Obviously, we do not do this work alone. Our colleagues in the core Scottish Government continue to deliver the systems and processes that we need to administer our benefits and work with ministers on the policy of devolved benefits. As always, therefore, I will not be able to comment on policy decisions.

While we have grown and changed as an organisation, I am also pleased to see some consistency both in the commitment of our staff and, most important, in the standard of service that we offer to clients. Social Security Scotland saw the second highest engagement score of any civil service organisation in the 2021 people survey, and I was heartened to note that, in our latest client survey, 94 per cent of respondents, when asked, said that they were treated with kindness.

Of course, much remains to be done. We are just past the midway point of the biggest agile delivery programme in the history of Scottish devolution, and our systems and processes are still developing. I welcome Audit Scotland commenting in its recent report on the focus on continuous improvement, which is very much at the heart of our operation.

With the Scottish Fiscal Commission forecasting that Social Security Scotland will be responsible for over £5 billion of benefit expenditure in the next financial year, including direct support to over 1 million clients, the importance of our work will only grow. I look forward to answering the committee’s questions.

Thank you. We will move on to questions, starting with Pam Duncan-Glancy.

Good morning, and thank you for your opening statement, Mr Wallace. We have heard a bit this morning about data collection, but Audit Scotland says in its report that there are inherent uncertainties about staffing. Will you explain what they are and what you are doing on staffing? Do you expect some of the spending reductions that were announced recently to have an impact on the staffing levels in Social Security Scotland?

On your first question, the uncertainties are as they have always been. I have read the Scottish Parliament information centre briefing and have reflected back on my appearance last year, and the uncertainties remain the same. Clearly, policies and systems behind benefits are still being developed before they come to us for implementation, so there will always be variables and uncertainties until we know what the system is going to look like. That is the main driver around the staffing level of the organisation. The more technology does, the fewer staff we need. The more manual processes we have, the more staff we need. Fundamentally, that will always be the case.

On the budget, I believe that we are in a good, strong position in terms of the administration costs of the organisation. Janet Richardson might want to say more, but we are in a good position in relation to the benefits that are close to being implemented, such as the winter heating payment. With the replacement for carers allowance, we are bit further away, and we still need to work with colleagues on exactly what that will look like before we can say what the staffing will look like.

I appreciate that.

We heard from the Scottish Fiscal Commission some information about data collection. What plans do you have to collect data on average payments and the payment flows? I know that there will be a transition period but, going into next year, what plans do you have to collect that data? Do you plan to publish it?

There are probably two points for me to make on that. I will come to the point about the Fiscal Commission last. We have started publishing our series of statistical information, which happens for all benefits. We will continue to do that for the adult disability and child disability payments on a three-monthly cycle. We have always planned to do that. That gives a depth of information on application volumes, the processing times and so on.

The Fiscal Commission has been looking for some quite specific information to assist with its forecasting. If you want, I will say a little more about that. We have been working really closely with the Fiscal Commission. Last week, we started some of the reporting that it wants to see. We draw the information from our systems and quality assess it before we give it to the Fiscal Commission.

Just to be absolutely clear, I note that it has never been the case that we sit on information that we have not shared with the Fiscal Commission. It is asking for information that we are not seeing, so the process has involved going back, working with our social security programme colleagues and coding the system in such a way that we can extract what the Fiscal Commission is looking to see.

There are probably two slightly different levels. I suspect that some of the information that the Fiscal Commission will want to see will not necessarily be appropriate for our statistical publication, so there will be a twin-track approach.

At the moment, is the Fiscal Commission relying mainly on data that is published or data that you are making available to it privately?

It is a bit of both. Up to now, it has predominantly been the published information. I caught a little of Graeme Roy’s evidence to the committee and I would echo some of what he said. We are in the early days. On adult disability, we just do not have any of the depth of data. It is not even a question of trying to find specific data that the Fiscal Commission is looking for. The system is in such an early state that we do not yet know what the flows will look like.

We are working with the Fiscal Commission on what it needs. In addition, we have been trying to give it a little more of the soft intelligence—which, in the absence of some of those hard figures, it has been quite keen to understand. My colleagues who are sitting on either side of me have been involved in some rounds of forecasting with the Fiscal Commission. That can involve some of the behavioural stuff and even some of our marketing and communications activity. It can find that information quite valuable. Not all of it is about hard statistics, although there will inevitably be some of those as well.

Did you do any work initially with the DWP to look at the data that it gathers to give intelligence not just for its operations but for the future of forecasting? How much of that did you consider replicating? If you did not replicate it, why did you make those choices?

Our analysts have always worked quite closely with the DWP. As with everything else, we have been working through our agile delivery. As I said in my opening statement, this is the biggest agile delivery that the Scottish Government has done since devolution. There is always a balance between paying to citizens money that they need and having a fully-fledged system in the background that does all the things that we would like to see. The data collection for the Fiscal Commission is just one of those things. From an operational perspective, Janet Richardson would like to see lots of suites of information as well.

It is not the case that this has surprised us or that we have ignored something. It is an inevitable consequence of the way that we are delivering. I do not mean that in any sense as a form of criticism. It is a feature of the agile delivery system, like some other elements that we might move on to talk about, such as technical debt. We have never been in a position to say that we will have everything that we need in place and we will then start delivering benefits. That is simply not the approach that we have taken.

Do you know when those systems will have the information that you need to collect built into them?

Again, there are two points for me to make on that. First, there is a tactical solution to that, and tactical work on it has already begun. We had a successful deployment last week, which let us see some of the reports, and our analysts are quality assuring that data before we can give it to the Fiscal Commission.

Secondly, there is a further bit of work about replicating what is available from the DWP, which you mentioned. The DWP has a mature and advanced statistical system whereby interested users can manipulate information directly. That is a much more strategic and longer-term bit of work. With our programme colleagues, we have set up a user group that is looking at the longer-term use of stats.

We have undertaken to try to get what information we can to the Fiscal Commission for the next round of forecasting. That work is well under way. Replicating, as it were, what the DWP has in place will be a longer-term bit of work.

Jeremy Balfour has a supplementary question on that topic.

Good morning. It is good to see you again.

I am a wee bit confused, given that the system has been designed from scratch. I appreciate that it was designed by the Scottish Government, which you now work with, but it seems to me, as an amateur information technology person, that we would want to start by getting the same data that the DWP collects so that we can compare apples with apples.

You are saying to us that we are not going to be able to compare what would have happened if people had stayed on PIP with what will happen with ADP because you are not recording it in the same way. From a scrutiny panel perspective, how do we know that the information will help to answer our questions? We will not be comparing the same things. I do not see why we did not design the system from the start so that we could compare the same things.

I am sorry if I have not explained that correctly. At a basic level, information is recorded on the numbers of people who apply, the levels of awards that they get and what conditions they get them for. The point about the agile development is that we do not have a sophisticated bit of software that we can put all that data into that will allow external users to manipulate it in the way that they want. We have not had that functionality in the system from day 1.

The systems are different, so they are not replicated exactly. The DWP will collect data on, for example, the outsourced assessments and the time that they take to come back in. Those things simply do not exist in our system, so we will never replicate that, because the processes are different.

We will of course be able to monitor and compare data on the levels of awards, how many people get which elements of awards and how many people take up the Motability scheme. We have been able to do some very early comparison. One thing that the Fiscal Commission is looking for is point-in-time data on awards almost in real time. We simply do not have the functionality to give that at the moment. However, there are clear comparables. We want and need to be able to compare the systems in terms of outcomes, but there will be some differences in what we monitor.

I have some questions about the processing times for the Scottish child payment. What is the current processing time for applications received since—[Inaudible.]

I did not catch the end of that, but I think that it was about the current processing time for the Scottish child payment. Is that correct?

Pretty much. What has the time been since the benefit was extended on 14 November?

It is too early to say what it has been since the benefit was extended—in particular, to clients with children up to the age of 16. We worked closely with our Scottish Government and programme colleagues to extend the benefit, and it should not be underestimated how significant a change and a technology advance that was. The Scottish child payment extension for the first time brings with it a degree of automation and straight-through processing, so a number of the 123,000 applications that I mentioned in my opening statement will effectively run straight through the system. Since we have brought those into the organisation, we expect the processing time for them to be very short. However, those cases that fall out will require manual intervention by some of Janet Richardson’s team. Therefore, there is a wide range of what a typical processing time looks like at the moment. The statistical publication that we publish routinely will set that out. The next one is due early in the new year, and that will be the first reflection of what the November intake has done for us.

10:15  

I will bring Janet Richardson in to say a little bit about how we are doing that and what that feels like, but it is worth saying, before I do, that this is the biggest intake of applications that the organisation has ever had. You may recall that, when we initially launched the Scottish child payment, we had a mechanism that allowed people to apply slightly early so that we could get our hands on the application in preparation for it. This time round, we have not done that, despite the numbers being more significant. We needed to rely on that straight-through processing to have a significant chunk of those applications go straight through the system.

Good morning, convener and members. In addition to what David Wallace has said, I note that more than 80,000 of the applications came in within the first two days, which shows what a huge number of people are waiting for this benefit. We do not underestimate that. We are doing as much as we can to process them quickly. Tens of thousands of people will receive that benefit before Christmas, but some will not. We have made it clear to clients that we will do our best, but we cannot guarantee that. We are keeping clients updated on what we are doing by text and social media, but, as David Wallace said, we will be more able to give a processing time figure early in the new year.

Is there any obvious way for you to improve the processing times or to help clients to track progress on their applications? For example, it has been mentioned that an online app or account might help with that. Would that improve things? If so, would you implement it?

My personal view is that automation is the way to improve some of the processing times. Continuous improvement on automation is, absolutely, a theme that we will come back to. The extension of the Scottish child payment is the first time that we have had the internal automation of applications. We are working with programme colleagues on the client-facing portal as well. My personal opinion is that that would assist at least in some way. It would let people who are able to do so to track their application online and get a sense of where it is. Inevitably, if they are able to do that, it reduces the demand on Janet Richardson’s front-facing contact centres and enables us to process applications quicker.

However, as always with everything that we do with social security, our programme colleagues will do some work on that. We have some feedback from the client survey, but we will do a little bit more research as to whether that would help us as well. My personal view is that people are used to seeing automated progress tracking, and those who are technically able to do that would probably find it helpful. We will see what users tell us about it. Again, I should absolutely emphasise that, from our perspective, the programme of delivery work for benefits is incredibly pressurised and incredibly tight. A new bit of technology would need to find a way into that timetable somehow. We do not have a separate, magic way of producing that to the side of the main delivery of benefits.

It is coming up to Christmas—maybe Santa can help. Thanks very much for that response.

Let us hope that Santa can deliver.

Having online capability to check was probably one of the few things that people liked about the universal credit system. Is there a reason why you did not build that in from the start, or does it go back to the agile methodology problem?

It goes back to the agile methodology. I agree with you. It has come through for a couple of years in the client survey that people who are comfortable and familiar online find that useful. They experience it in other places as well. They are used to it when dealing with commercial utilities and others, so it is a logical thing for us to explore. You are absolutely right in saying that the reason why we do not have it comes back to the same point about statistics, the things that we have been operating and what is called a minimum viable product. We have been layering this up as benefits have come on board. We have had four years, in essence, of minimum viable products being built up. Every time we do that, more functionality comes into place, but this has never been a place of developing one system and then running a benefit live as well.

I could say a little bit more about that if it would be helpful, because another thing that is different from the DWP is that we have one base system underpinning all our benefits. We do not build a system for the Scottish child payment or a system for best start grants and add it into the organisation. There is a single platform. Every time we drop in more functionality, the testing of that platform becomes more difficult and complex. The amount of time that we have to do that without bringing down operations and closing the system to the public becomes more pressurised, so we try to do that over a weekend period. All of this is a balance. Do we want to wait for clients to be able to see progress of their applications online or do we extend the Scottish child payment? We have to make some of these choices.

I understand that. I can see that, because of the choices that have been made so far. I get that there are countless problems with the DWP and the way in which it does things, not least policy. Can you explain why it was decided not to just lift the technology that the DWP used and apply it to the policy and the specifics for Scotland, instead of completely starting from scratch and facing the situation that you have just described, which is having to shut down the system for the weekend to try to catch up, inevitably continuing to have to bolt on additional bits to IT and having what appears to be a substandard system?

I disagree with the point that it is substandard. There are pros and cons to it. The DWP is not a single system. There is no single system that we could have looked at and said, “If we take that, that will support all our benefit deliveries.” This is a challenge that we come to when we talk about data with the DWP. You are scanning lots of different systems, some of them quite old and some of them going through transformation programmes. It was not as simple as saying, “If we could lift or replicate this thing, we could bring it here and it has all the functionality attached to it.” The DWP’s systems have grown up over decades in different ways for different benefits. There was not an option to take one thing at the core and replicate it.

We have ended up with a primary contractor, IBM, that works with the DWP. Again, at the heart of this is a bit of technology that works in elements of the DWP. We have gained advantages of understanding how that replicates. There was simply never a thing that said, “Take system X and bring it up and replicate it.” We have had both benefits and disadvantages of that not being available. These options were looked at in the early stages of the programme and there simply was not a thing to lift and shift.

That is helpful to understand—thank you. My final question on this is about the fact that 200,000 applications for the Scottish child payment were expected when the increase came in but, in the event, there were only 89,000. What happened to the other people? What information do you have about those missing people? How much of it is to do with the fact that the website either was undergoing some development or crashed? Have you looked into that?

I will take the numbers question first. The figure of 200,000 was for children—it is the Scottish Fiscal Commission’s figure for the number of children who were expected to benefit. The figure of 123,000 was for applications. Some applications will be for more than one child, so we will not know how many children those 123,000 applications represent until we have some further statistics through the system. To give you some illustration, until we had extended the Scottish child payment, the ratio was approximately 1.3:1. That meant that for every application we had about 1.3 children. Naturally, we expect that to grow a bit, given that the age range has increased significantly. Within those 123,000 applications there will be more than 123,000 children. We need to understand, if there is a gap, what that gap looks like. Again, we are in the early days and we need that to flow through the system before we have that information.

I have a couple of questions. One of the measures in the charter is the percentage of applications processed within 10 working days. In 2021-22, the majority of claims were not processed in that timeframe. Why is that the case? Do you hope to do better next year? If so, what target are you looking to set yourselves for the coming year?

As you know, we have always been quite careful about setting a target, mainly because it could drive the wrong type of behaviour. The 10 days in the customer charter is quite an ambitious measure. It is right that we measure it, but it was always going to be a relatively small proportion of cases that would hit that 10-day figure. The thing about the customer charter is the balance of all those measures—how people are treated, the outcome and all the rest of it. The 10 days will be incredibly challenging, particularly for disability-related benefits. For low-income benefits—I will bring Janet Richardson in on this—the automation that we have referred to in terms of straight-through processing and having a single application for multiple benefits will help to drive down the processing time.

When Janet Richardson was in front of the committee supporting the minister previously, we were having a period during the summer when our processing time was too long. Janet covered some of the activity that we are trying to do to drive that down. It will be continuous improvement of the organisation, automation and a focus on some of these things. The 10 days was never a target that we would try to work to. It is right that it is in the charter and it is important, but it is one of those things that we need to balance off with the other measures, such as the 94 per cent measure.

If you have a charter, it is uncommon to say, “We will never hit this figure,” or, “We will not get most of it done.” What is a realistic time for an application? If I put an application in for ADP, what is my expectation?

We are doing a bit of work at the moment with clients on exactly that. I know that we faced some criticism when we took the eight to 10 weeks indicator down from the website. The reason we did that was so that we did not mislead clients as to what was expected. Ten days will not be achievable for disability benefits. Not least, we have changed the system. We are in a position now where we collect information on behalf of clients. I cannot see how we will get to a position where the vast majority of disability benefits would be achievable in a 10-day turnaround. We work with clients now on exactly what those expectations are and also how best to realise those expectations. Is it a static website indicator? Do they want something more dynamic? Do they want texts giving a more regular update of where we are with processing? A number of measures in the charter balance out. I certainly do not dismiss the importance of making timely decisions. We saw over the summer that we were getting to a level where both individual clients and stakeholders were unhappy with that service.

As a committee, we need to go back and discuss that with the Scottish Government. If we have a charter and people read that charter but we will never hit those figures, it is almost not worth the paper it is written on.

10:30  

Back in October, your colleague Janet Richardson came and gave evidence. She said:

“We recognise that, with applications opening so near to Christmas, people will be concerned about getting their money by then.”—[Official Report, Social Justice and Social Security Committee, 27 October 2022; c 6.]

Clearly, this was an issue that you were thinking about back in the autumn. On Facebook yesterday, you put up this comment to your clients:

“Because of the volume of applications, some people who applied in the week may not get a decision until early 2023.”

What discussions did you have with the Scottish Government to say, “We cannot deliver this for a substantial number of people”? What planning did you do so that we could limit the number? The number of people who will not get their money seems fairly high. We are trying to devise the system that we have all talked about with dignity, respect and so on, but we do not seem to be doing that with this new system.

I will bring Janet Richardson back in to talk about the expectations and communications. I come back to the point that the extension of the Scottish child payment to those under 16 within the existing devolution programme is the biggest single intake of applicants that we will ever have into the system. How we transition that volume in at a single point in time will always be enormously challenging. We cannot physically gear our staff up to give the same level of service on week 1 or 2 as we will once that becomes a more business as usual activity. These conversations are had routinely with the Scottish Government. In the way we develop our systems and our processes, as part of the agile delivery, we have interdisciplinary teams. Our staff work with the Scottish Government constantly on the implementation of systems, what this means and the impact on the organisation.

We have also become a large and complex organisation now. You point to the communication. The communications that we put out drive demand into the system. We will have known that that Facebook post would bring people into the organisation contacting us. These things are done through the programme and through that delivery chain as part of an integrated team. When I say that we are having discussions with the Scottish Government, I do not mean that we have an exchange of letters saying, “This is what the impact will be.” We are working with the Scottish Government on it. Coping with that huge surge in demand in the first week was always going to be a huge challenge for us. We did not do what we did at the outset, which was take applications early. We did not do that, because we needed the automation of the function to work and to push cases through. Janet Richardson might want to come in to make the point that we need a level of automation for those cases to come through.

Yes, we do need that level of automation. The other reason why we chose to do the launch in this way is so that some people would get their money before Christmas. That was the earliest day under the legislation that we could open for applications. Rather than saying, “Apply now,” and paying later, we knew that it was important for some people to get their money before Christmas. I do not think that we ever said that everyone would. We said that we realised that people would like that to happen. Part of what we do is manage people’s expectations. That particular Facebook post yesterday was to make sure that we could do that. We told clients that they would get their benefits before Christmas, but we did not want them to assume that that would include the increase of Scottish child payment that they had recently applied for. It is a balance between managing clients’ expectations, making sure they get the money that they need and letting them know that payments will be backdated. I accept that, at this time of year, people would rather have the money now, but we never promised that we would be able to deliver that to everyone this side of Christmas.

Thank you.

Miles Briggs will take us into our next theme.

Good morning, and thanks for joining us today. I want to touch on the charter research, specifically with regard to improvements that the organisation is taking forward following that research. It is worth saying that, by and large, the findings were positive. What learning have you taken from that?

By and large, the picture is positive—I do not say that in any way to be complacent. We need to remember that the charter research reflects the financial year 2021-22, so it essentially speaks to a group of clients who have not gone through the ADP process. As with all these things, it shows a point in time.

Coming back to some of the previous questions, the continuous improvement element concerns the timescales. We are getting a clear indicator now that timescales—particularly at that point in time—had reached a point at which people found them unacceptable, and we needed to improve that. It ties back to the communications point—how we communicate with a broad range of clients and whether we can do some things smarter with technology, as well as whether we need to focus on doing things slightly differently with particular groups.

We have talked a lot about technology, and I agree with the technology point. Technology will allow us to focus our scarce resource on those who most need it if the majority of people can work through a system digitally, by choice. I might bring Janet Richardson in to talk about the working group on that communications element, but the thing that I personally took from the charter research goes back to the point that has been made about timelines. We need to focus on the continuous improvement in the organisation and focus on timelines. I might come back in and say a little bit more, but I will let Janet Richardson come in on that.

As David Wallace said, probably the key thing for us is that processing time—in the disability benefit space, in particular—as much as anything. It is important to note that, among all the case managers we have recruited and trained and are delivering, we still have some who are not fully in that live service yet—and they will not be until the end of January. We will start to see things progress and ramp up from the new year. In addition to that, we have talked about how it is better for us to keep people informed and updated while we are still working through their claims.

An area that we talked about was how we communicate with older people in our communities. Our local delivery teams are out and about doing a lot of work. They do a lot of cost of living events and warm hub events with stakeholders. A lot of the work is not just about improving our service, but is about improving the contact we have with clients to make sure that what we deliver meets their needs.

That is helpful. Thank you. On that point, how have you taken forward the signposting towards the VoiceAbility advocacy service, specifically for non-verbal clients, since we last met? There was a lot of talk about recruitment of the required individuals.

In general, we have worked closely with our VoiceAbility clients. VoiceAbility has also come to talk to the stakeholder group that we meet with regularly to make sure that not just us but stakeholders are aware of the impact that VoiceAbility can have and the services that it delivers. They have taken away some actions to work closely with a couple of those groups. We also invited VoiceAbility into the organisation to spend time with our client advisers, not just the ones concerned with local delivery, but the ones concerned with telephony as well, to make sure that they are aware of the service. We have also appointed advocacy champions within the organisation, who regularly keep new client advisers and teams up to date and remind them that they need to ensure that clients are aware that there is an alternative to what we offer in terms of local delivery.

I have met VoiceAbility’s chief executive, Jonathan Senker, on a couple of occasions. On much of the work that Janet Richardson is outlining, we have been keen to see the organisations working together. Clearly, we also need to do it in a way that protects VoiceAbility’s independence. A lot of what we have been trying to do is link it up with stakeholders who were already working with us, through things like the stakeholder forums that Janet Richardson mentioned. For us, it is about that advice for our clients and their awareness of it. We want to increase that awareness of the service within our client advisers as well.

This is not so much a question, but it would be helpful for the committee to know the data around where those individuals will be based. That has not been clear. Also, what is the potential for home visits? Originally, that was flagged as one of the key reforms, but I have not been able to find out whether those have taken place or how many people have applied for them. I do not know whether that is within the systems data that you can provide.

VoiceAbility is doing some stuff and we are doing some stuff. To ensure the independence that I mentioned, we do not manage the VoiceAbility contract—we did not let it and we do not manage it—so that is probably a point for Scottish ministers and policy colleagues to address. We are not sighted on and have no input into where VoiceAbility puts its staff or how those appointments are delivered.

We now have local delivery staff in all local authority areas in Scotland. With the adult disability payment, their activity has been ramping up. We can probably give some of the data about visits that they are undertaking, but I can absolutely confirm that those home visits are happening where they need to happen. We have strong examples of some of our local delivery client advisers repeatedly going out to do a home visit if they feel that they need to provide additional support. It is an important difference in the system. We talk about PIP moving to ADP. I always focus on clinical assessments changing to be consultations internally, the application process and local delivery. Providing support where people need it is important. The Government has invested in SSS to resource up to enable that to happen. In some ways, that should reduce the need for advocacy in some places, because we provide people with help with their applications when they need it.

I have a quick question about redetermination for the CDP. Correct me if I am wrong, but, as I read the figures, 85 per cent of requests for redetermination were granted. Why do you think that is such a high rate? It is certainly higher than it should be under the DWP scheme. What is happening with that first application by your staff?

I will say a couple of things and let Janet Richardson say a bit more, as she manages the redetermination teams. Again, I caution slightly against a direct comparison with the DWP. We have always said that redetermination should be done by people who have never been involved in the case before and can take a step back and look at it afresh. Janet Richardson will say more about this, but I think that the high level is probably driven by additional information coming in. I do not have figures to hand about what percentage is driven by additional information, but there will be a significantly high volume of cases in which we gather more information at the redetermination stage than was available to the people who made that first decision.

There is also probably something about the newness of the process. For example, in their early days, low-income benefits would tend to have a higher redetermination level. The organisation is new to this. We have a new scheme, a new policy and new staff. Everything that the staff touch is effectively being touched for the first time at this point. Predominantly, that figure is driven by new evidence coming in at the redetermination stage.

One of the changes that was meant to happen under this scheme was that you would look for the information the first time around. Clearly, that is not happening. What training is now being given so that it is not the claimant providing that information but you finding the information to get it right? The application form and the way that people make an application is different, but the criteria for CDP have not changed at all. It is a new system but they are not new criteria—people are still getting an award on the same criteria. Again, has the training not been there? Why have they not gone for that information the first time around?

I will let Janet Richardson say a little bit more about that. You are right in saying that that is part of the system, but there is also the newness of everything else in the system. There are changes to the system—changes in terms of being able to draw on the clinical expertise that we now have in the organisation, which does not exist in the DWP system. The point about newness—sorry if I did not make it well—was about the newness of the organisation. We have roughly doubled in size every year that the organisation has existed. The people who are administering CDP and the adult disability payment are new to the organisation. They have almost exclusively joined within the past year and during a period when Covid restrictions might have been in place. Inevitably, as we go on, people will become more experienced, more efficient and more productive.

We seek to get evidence up front. I will let Janet Richardson say how that happens at a practical level.

10:45  

When we talk about the information, we are not necessarily talking about information that we need to gather; sometimes it is the information that the clients put on the application forms. We realised early on with CDP that, for all the right reasons, we had made a lot of the application form voluntary with regard to whether the client gave us that evidence. We realised fairly quickly, because of the number of redeterminations that were coming in, that the client could have told us a lot more, which would have helped us to make a different decision. It is fair to say that quite a significant number of that 84 or 85 per cent of clients were receiving the benefit—their redetermination was about the award rate. We realised that we needed to ensure that the clients realised that the more detail that they can give us about the impact their condition or their disability has, the better, and we changed the wording, particularly for the online application, to reflect that. Of course, we still gather that supporting information on behalf of the client as well.

Since then, we have also put in place a step whereby we contact the client for both adult and child disability payments when they make that claim to talk to them about whether there is anything else they can tell us. A good example of that is the fact that we can spend weeks waiting for a general practice surgery to send us a list of the medication that a client is on, whereas a client often has that to hand because they have their repeat prescription and could upload that information and send it to us. It is not necessarily about our not looking for that information—we do—it is about how we gather it. We have learned some positive things from those early stages in relation to CDP.

I have no further questions. Jeremy Balfour’s last question, which has just been answered, is the one that I was going to ask.

In that case, I will move back to Jeremy Balfour.

You may not be able to answer this, but I am interested in the matter. The latest figure that we have is that, under ADP, there were 1,845 new applications out of 3,545. That seems quite a high figure. Do you have any early indication of why there are so many new applications at this stage, or is that what you expected when you went live?

I do not have those figures in front of me—Janet Richardson might come back to that. I saw a little bit of your earlier session today with the Scottish Fiscal Commission and heard what was said about the overall increase in applications across the UK. What has been seen at a PIP level in the UK we definitely see in a Scottish context in terms of increased applications. Anecdotally, we heard that, in the welfare advice and benefits advice sector, there had been applications stored up. People had deliberately not been making an application under the DWP system if they were close to a pilot area or if their application would be made close to when the national scheme would go live in Scotland. Therefore, those figures are not out of kilter with what we anticipated. In fact, if anything, we saw a range. We planned this on a high and a low range in terms of potential applications. The adult disability payment has more or less come in through the middle of that expected range. Janet Richardson might have those numbers to hand.

I do not have them to hand, my apologies, Mr Balfour. I did not understand the question until I listened to what David Wallace just said. We saw that spike initially but, when we look at the head of work, it is pretty much where we would have expected it to be in what we call our average forecast for now—it is pretty much tracking that week in, week out. The initial spike looked higher than we had anticipated, but, in terms of where we thought we would be now, it is about average.

That is helpful. Earlier, you spoke about work that you had done in the summer on when people should expect to hear from you about both an initial decision and a redetermination. If I apply, what is my expectation now for hearing about a first decision? Then, if I have to go to redetermination, how long that will take?

From the most recently published statistics, it takes in the region of 54 working days for a decision to be made. As is reflected in the SPICe figures, that is on a bit of an upwards trajectory. That 54 days is based on the last month of that data being about 77 days. We expect that to stabilise and ideally come down, but that is the current working time that we are looking at.

My final question around this is a parochial one concerning the Lothians. The figures show that 55 people have been transferred from PIP to the new ADP—I am not one of those 55. Is that the speed of the process you expect? It seems quite slow. Do you have at least a working idea of when the full transfer will take place? Are you reaching the targets that you already had? Are you confident that everybody who is now on PIP will be transferred to ADP on the timescales that we discussed previously?

Yes. The challenge with those published figures, Mr Balfour, is that they were from the start of case transfer. That was, as you have alluded, a deliberately slow start so that we could check cases as they came in, check that we were getting the data and information we expected and check that the process was working. Those numbers have already significantly ramped up, and there is a plan to transfer all those cases between now and the end of 2024. We see nothing at the moment to suggest that that is not achievable. At points, it relies on quite significant volumes coming in in a single month, and we want to check that we are able to manage those big capacities. You quoted 55 transferrals in a particular month. There will be a month in which we get to somewhere in the order of 30,000, and we want to make sure that we can deal with that.

Essentially, the lag in the data is the problem with that particular statistic.

Theme 6 is on error and fraud. I will bring in Pam Duncan-Glancy.

When the system was set up in Scotland, the Parliament at the time said that it would be fairer. I am slightly concerned about some of the work that I have seen on counterfraud and surveillance. Can you tell us a bit more about the team that you have set up to do that? What sorts of things are you doing? The job description included things such as covert surveillance, which made me quite uncomfortable. Could you tell me a little bit about that? Also, how much are you spending on the counterfraud surveillance work?

I will bring in James Wallace, who leads that counterfraud team, to give a bit more detail.

I want to give some assurance on what that feels like for clients in the system. To be absolutely clear, we will have fraud in the Scottish system—that is undoubtedly the case. Any system of social security will have fraud in it. Our rationale for having counterfraud is to limit and detect that. The surveillance function and the ability to counter fraud are designed as a disincentive. The biggest element in that will be organised fraud, and I will let James Wallace say a bit more about how we tackle that.

It is not about criminalising our clients in any way, shape or form. You will know that, during the passage of the Social Security (Scotland) Bill, there was robust debate about error versus fraud. I separate error and fraud, as the committee has done. Fraud is quite specific and different, but it exists and we need to tackle it. We have the full range of powers, including on covert surveillance. That will be used incredibly sparingly.

So that I do not say anything inappropriate, I will let James Wallace come in and say a little more about the set-up of the team and, as far as he can, the cost of the team and what it is currently doing.

I had a suspicion that we would cover this issue today. In any benefit system, you will encounter fraud—it will be a factor. My focus is organised crime that will attack our benefit system. That is a definitive risk in the system. We have developed counterfraud capabilities over the past number of years and we have a suite of responses available to detect, prevent and address any fraud that we find in the system. We do not release specific detail on the cost of the team, because that would allow bad actors in the fraud space to size the team and our fraud response. We are careful about what we release. That is not an attempt to be secretive or to lack transparency; it is so that we do not undermine our counterfraud efforts.

Specifically on covert surveillance, my main point is that that is a highly regulated area. There are statutory powers that we would use when employing covert surveillance. The Regulation of Investigatory Powers (Scotland) Act 2000 is the legislation that gives us our covert surveillance powers. The legislation sets out areas where we can use those powers and tests for that. The powers are not used lightly and are highly controlled. Social Security Scotland has a code of practice for investigations, which we published in February 2020 and is available on our website. It sets out for members or for clients exactly how we use those covert surveillance powers.

It is a tiny minority of actors in the system who are seeking to defraud us, and covert surveillance is used only when it is legal, justified and proportionate—frankly, it is a last resort when our other tactics have failed. It happens when there is significant theft from the public purse. It is not designed to covertly surveil an individual who has claimed the best start grant fraudulently—the powers would never be used for that. Indeed, the powers and capabilities of the team were set up to be available and developed for the launch of the larger and more costly benefits to the public purse, such as disability benefits.

If an industrial-scale fraud was perpetrated against the organisation relating to disability benefits, which are a high-value and regular payment, the losses to the public purse could be significant. It would then probably be appropriate to use all our available tactics to counter that fraud, but, if they were exhausted, we might move to covert surveillance if that was entirely necessary, legal and proportionate.

I appreciate that. Can you describe the sort of organised crime that you are talking about?

It is when there is a co-ordinated deliberate attack on the benefits system by someone who is not eligible for benefits. They make deliberate misrepresentations to steal from the public purse. Those are the people we are talking about. We are not talking about an individual who inadvertently makes an error on a form or who ticks a box or even misrepresents themselves on a single occasion for a single benefit. Those are not the main threats to the system. It is about organised crime.

Under the recently published Scottish budget, the scale of public money for which we will be responsible in 2023-24 is £5.1 billion. That is a significant element of public money. Unfortunately, fraud is a factor in our system. It is a tiny minority of individuals. I do not even call them clients, because they are not eligible for our benefits—that is what makes it fraud. People out in society will seek to attack the public purse. The scale of money that is involved requires us to have powers that are commensurate with the response that is necessary to protect the public purse. Any pound that we save from a criminal’s bank account can go to an eligible person or another Scottish Government policy imperative. That is what is behind our counterfraud approach. I recognise that the powers are strong, and I assure you that they are not used lightly.

What evidence do you have of that kind of fraud and the likelihood of it happening?

11:00  

The DWP has had its accounts qualified for the past 30-something years due to irregular payments relating to fraud and error. That is a long-standing factor of the benefits system. The levels of fraud are small. We generally talk about 1 per cent of the benefit expenditure, but 1 per cent of benefit expenditure is a lot of money. At UK level, it could be £1 billion. At the Scottish level, it could be £50 million. It is a significant sum of public money that we ought to take steps to protect. There are not that many organised crime groups that we are looking to guard against, but we must remain vigilant or we will lose money.

I will bring in Miles Briggs to ask a supplementary question and then to lead us on to the next set of questions.

My supplementary is linked to that point on threats to the system. I noticed that the national cybersecurity centre has been offering expert guidance to the organisation. Have you had any cybersecurity breaches to date? What strategies are in place to build in systems to prevent that, especially given what we have seen recently in organisations such as the Scottish Environment Protection Agency?

I will answer that question as best I can, but I caveat my answer by saying that I am not technically minded. Our chief digital officer would be happy to come and give more in-depth detail on that. We worked with the national cybersecurity centre at one point and, since we launched, we have embedded and built up expertise. Ever since we have had a digital presence, we have had teams monitoring that. There will always be attempts to breach the security, whether it is by bots or by people in foreign places. That happens absolutely routinely.

I am not conscious of any particular cybersecurity incidents that I want to bring to the committee’s attention, but we are never complacent about that. It is a bit like the organised fraud point that James Wallace outlined. We know that attempts will be made on any system that makes money available online. As soon as you put up a website, you will find that it automatically comes under an element of testing attack, as it were.

That is helpful. Given that a new system is being put in place, it is important to build in that approach from the outset. I believe that the Deputy First Minister has now put in place a reporting function for all public agencies to report back to the Scottish Government on that.

The theme that I want to ask about relates to new benefits and improving systems. What scale of technical debt is there currently and what progress is being made to develop a structured approach to address that?

Our chief digital officer has led on putting in place a framework for capturing technical debt. I should say from the outset that technical debt, as I think is identified in the Audit Scotland report, is a consequence of the agile approach. We would not look at technical debt and say that something should have been done but has not been done. It is a consequence of the approach of having minimum viable products and focusing on delivery to citizens. Inevitably, some functionality—some of the stuff that we have mentioned already such as a statistics function and an online portal for clients—has not been achieved. We want those in an ideal world, but we do not yet have them. The chief digital officer is currently capturing all that information and assessing it with our programme colleagues.

I will turn to James Wallace on the financial elements. The capture process that I referred to is about functionality. I do not know whether James wants to say how or when that might be translated into financial figures.

The answer is that that will happen over time. We have a standardised approach to assessing investment decisions in the agency. We would always consider how much any proposal to service a particular piece of technical debt will cost and how much it will pay back. There is a value-for-money assessment of servicing a piece of technical debt. A process will need to be gone through once the initial process of devolution is complete to analyse what we wish to fix, where the benefits are, what it will cost and where we might not wish to service it because it does not pay back and is not good value for the public purse. Those arrangements are fairly standard in government and are already in place. The agency will move into a cycle of continuous improvement whereby we make investments that are of value.

That is helpful.

You have answered some of the questions on the need for additional data. Do you have timescales for when you think measures will be put in place to draw down that data or have datasets that can be used?

In relation to the Scottish Fiscal Commission’s specific asks, we are trying to do that for its next round of forecasting. That work started last week. We will start to see what those reports look like, and our analysts are then quality assuring. Our commitment to the Fiscal Commission is to try to do that for the next round of forecasting. As I said, the longer-term work is with the community of stats users in Scotland. The Fiscal Commission will be one user, and some academics will be others. We will gather that stakeholder group and consider what the ask is. Again, it goes back to what can be achieved and when. That is a longer-term bit of work. With the Fiscal Commission’s asks, we are committed to trying to meet those for the next round of forecasting, which will be early in the new year.

That is helpful. Perhaps you could keep the committee updated on that, because we will have a watching brief.

Since the internal audit report in July, what sort of guidance and training has been undertaken? Where is your strategy or policy around working from home? We discussed that on our visit to the Dundee office. A number of people are in new buildings in Dundee, for example.

I will cover the training point and then bring in Janet Richardson on it, then I will go on to the question on hybrid working.

Guidance and training were highlighted in the Audit Scotland report because there was a particular risk in the run-up to the introduction of the adult disability payment. It is a feature of the way in which we deliver such benefits that guidance inevitably comes to us late. Again, that is not a criticism of anything; that is just a feature of the system that we have.

There is a real challenge in getting guidance to people. Bits might be missing from guidance that are not required for day 1. Redetermining guidance might not be there on day 1, because it is not seen as a day 1 function. We get into a cycle of continually training people and the guidance changing. Even as we train people, the system that they are training on does not necessarily reflect the live system. We have to put people through training on a system in one place with guidance, and, when they come out to consolidation, it will look different. When they come out into live, it may look slightly different as well. That is the risk that Audit Scotland was highlighting.

That risk was specific to the benefit launch of adult disability payment, and it has dissipated for now. It will come back when we go into the next big benefit launch—possibly less so for the winter heating payment, but certainly for the replacement of carer’s allowance, for example. We can make some improvements.

Janet Richardson might want to say how we do some of that consolidation training, particularly around guidance. The nature of what we do means that guidance will continually change, and that is a challenge for the organisation. There is a continual challenge of moving guidance, and we have to train people all the time on that. Janet, do you want to briefly say something on the training aspect?

You have probably covered most of it. We learn as we go along. The more experienced we become in the agency, the more we are able to influence the design of the training from the beginning, which we were not able to do at first. A lot of our feedback from our previous training goes into that, and the situation is much more stable than it was.

If I have time, I will come back in on the hybrid point. When the committee visited the organisation, we were probably still trialling approaches. We have decided that we want to become a hybrid organisation. We are deliberately not moving to a fully remote working-from-home environment because we think that it is quite important, particularly given where we are in our development, to get our teams together for reasons of training, development and building and maintaining our culture.

We have said to all our colleagues that we are looking for a hybrid approach. People will fall into rough categories depending on how mobile they are and whether local delivery, for example, would look different for them. We have set a clear expectation that we want to see people in the office for at least two days a week. That ratio essentially means that our estate can support the workforce that we have. At the moment, we see about 50 per cent occupancy, and we want to drive that higher in the new year.

Hybrid is the approach that we have taken. We have been reasonably direct about saying that we want to see people in the office, and we are convinced, as a management team, that there are benefits to that. People who are in the office are now starting to see those benefits as well.

The minister told us that his expectation was that all clients would be paid winter heating payments in February. Is the system in place to make that happen?

Again, the just-in-time nature of the situation means that there will be an element of our needing to see the data and what drops out of that. There will be a challenge regarding how much manual intervention is required. I might briefly bring in Janet Richardson, because, as I said in our previous answer, we work hand in glove with our programme colleagues on that—it is a multidisciplinary team. We have built up resources based on what we think at the moment, and we think that we have the right resources, but there is still a requirement to see what the DWP data looks like.

We will get that data in January, so we will have a better idea then. As David Wallace said, we already have the resource that we think we will require in training, and we have additional resource that we can use to make sure that as many clients as possible get that payment in February. We are relying on getting the data in time to do that.

Our last questions will come from James Dornan.

It has been more than a year since the launch of your local delivery service. How is that going? What have you learned? How involved have the local delivery teams been in the continuous improvement theme?

I will say a couple of words, then I will ask Janet Richardson to reflect on that. You are right that it has been more than a year, but the adult disability payment is the element that has brought local delivery to the fore.

I am incredibly proud of the work that has been done in local delivery. As I mentioned earlier, clients who would not otherwise have completed an application form are being practically assisted to do so. One of our founding principles is to assist people to get the benefits that they are entitled to, and we are seeing local delivery actively working to do that.

The way in which that has been done is important. Of course, we have to be mindful of resource and there is an element of triage, but, where necessary, we have had people in people’s homes and we have had people on sparsely populated islands, taking ferries and putting their wellies on to get to the places where they need to be. We have seen local delivery leads engaging with local authorities and local voluntary organisations and becoming part of that support community. In some instances, I hope that we will have seen those leads engaging with local MSPs and MPs and getting engaged in surgeries and workshops.

Personally, I am proud. I have probably missed some of the important stuff that local delivery leads are doing, so I will let Janet Richardson have a quick word on that.

You have covered most of it, but it is important to highlight a couple of key things. The local delivery team has been supportive of people coming across from Ukraine. It has been working with people on the ships where Ukrainians are living, and we have received quite a number of claims that the team has taken on behalf of people from Ukraine, whom it has supported through the Scottish Government’s benefits system.

The team is also involved in pilots in certain areas, such as Dundee. We are looking proactively there at reducing child poverty. The team is linking in with Dundee City Council and the DWP in that area to work through cohorts of clients—whom perhaps all of us come into contact with—to make sure that they are receiving everything that they are entitled to. The local delivery team does much more than just go out and help people to fill in a claim form. It is an integral part of communities across Scotland.

Once again, I thank the panel for attending this morning and for giving us your evidence. We will now move into private session.

11:15 Meeting continued in private until 11:25.