Work and Pensions Committee 

Oral evidence: Safeguarding vulnerable claimants, HC 146

Wednesday 13 December 2023

Ordered by the House of Commons to be published on 13 December 2023.

Watch the meeting

Members present: Sir Stephen Timms, (Chair); Debbie Abrahams; Siobhan Baillie; Neil Coyle; David Linden; Nigel Mills; Selaine Saxby; Ben Spencer; Sir Desmond Swayne.

Questions 58 - 121

Witnesses

I: Joshua Reddaway, Director, Value for Money, National Audit Office; and Ashley McDougall, Director, Value for Money, National Audit Office.

II: Brian Dow, Deputy Chief Executive, Rethink Mental Illness; Daphne Hall, Vice-chair, National Association of Welfare Rights Advisors (NAWRA); and Jamie Thunder, Policy & Public Affairs Officer, Z2K.

Written evidence from witnesses:

• Rethink Mental Illness (SVC0064)
• National Association of Welfare Rights Advisors (NAWRA) (SVC0028)
• Z2K (SVC0052)

Examination of witnesses

Witnesses: Joshua Reddaway and Ashley McDougall.

Q58            Chair: Welcome, everybody, to this meeting of the Work and Pensions Select Committee and our inquiry on safeguarding vulnerable claimants. A very warm welcome to the two participants in our first panel. Please tell us very briefly who you are, starting with Ashley McDougall.

Ashley McDougall: My name is Ashley McDougall. I am the Director of Health Value for Money at the National Audit Office.

Joshua Reddaway: I am Joshua Reddaway. I am the Director of Fraud and Propriety at the National Audit Office. For six years, I led our work on the Department for Work and Pensions.

Q59            Chair: Welcome to both of you. I will put the first question to you. Departments define vulnerability in different ways. Do you think DWP could usefully apply a definition of vulnerable adults used elsewhere in government? Are there some good models that you think the DWP should be learning from?

Ashley McDougall: It is quite functionally related. It depends on what vulnerable means in context, what people are vulnerable to and what you are trying to protect against.

Joshua can talk about the DWP, but the work we did last year on safeguarding vulnerable adolescents was largely focused around the Department for Education and its safeguarding responsibilities for adolescents. The Department owns the policy but it is delivered through local councils and children’s social care services. It is about risk of violence, harm and danger, threats to life for children, so it is quite a high threshold. Beneath that, there are various other thresholds. It is quite specific. It is not the DWP assessment model.

We work to other definitions but that has been the principal one. MoJ looks at prisons. DCMS looks at children through youth clubs. Different Departments have different interests depending on their purpose for spending money. We decided that it was not productive for us to try to align the multiple definitions. There was effectively a family resemblance among the vulnerabilities. Departments were probably interested in the same population but were coming from different directions and it would not have helped to have a single definition, so I am not sure that there is a transferable, single-use definition.

Q60            Chair: Did you find that there was good communication between the Departments looking at these related things or were they all doing it on their own, in isolation?

Ashley McDougall: We found at a couple of levels that strategically it was not particularly joined up. I think we found 16 different committees with an interest in an aspect of vulnerable children and adolescents and it did not take them very far because it was not granular enough. Locally there were examples. For example, the MoJ worked with the DfE to share data looking at who was in school, who was not in school, what crimes had been committed, who had been sentenced and what was the overlap in those populations.

There were bilateral joins but no system joins across the systems. They talk at the top but there is no structured apparatus to pull all the information together and allow the Departments to see the gaps in the system as opposed to one Department talking to another, which is good as far as it goes but it is a bit limited.

Q61            Chair: When you looked at all this, was the DWP involved in the discussions, and collaborations and should it be? Should the DWP be more plugged into what is going on elsewhere across government?

Ashley McDougall: For the thresholds and purposes that we were looking at—risk to harm, risk for bad outcomes for children engaging with the criminal justice system or the children’s social care system—DWP did not seem to be a player and I am not sure that it needed to be.

The vertical line from central Government goes down through local government where a lot of the services are delivered locally—social care, education, the NHS, the police—and they will work together locally. We found that local working is not embedded yet and that has been a long-standing problem, trying to get all local information and interests together. The three statutory partners for that are the local authorities, the chief of police and the NHS ICBs. Where they come together through multi-agency safeguarding partnerships, they are still bedding in but they in turn were replacing something of joining groups that did not work before. DWP is not a party to those local discussions.

Q62            Chair: Maybe this is a question for you, Joshua. Do you think that DWP should be more plugged into these emerging arrangements?

Joshua Reddaway: I think you started with the right question—what are the definitions, what exactly are you trying to do? Sometimes we are at risk of not being clear about vulnerability to what and in exactly what way DWP relates to it. The reason I say that is because I think it depends on how DWP is operating and what it is trying to achieve. Can I tentatively suggest three different definitions that we see in our work? I am not sure if it is complete but it might be helpful.

The first is the one that Ashley talked about, which is vulnerability to harm. DWP relies on there being a functioning system in place to support people through care and the criminal justice system in the way that Ashley talked about. DWP does not see that as its core role and, therefore, has ways that interface with that rather than wanting to be a core player in it. You see, for example, work coaches who are often one of the first bits of the state that vulnerable people come into contact with because, unlike other bits of the state, DWP requires people to come to it to claim benefits. That means that work coaches and the 10-minute conversation that they have with somebody on the intensive work search might be the interaction that a person has with the state. During that time, the coach’s process is really around, “Do I need to refer this someone to somebody else?”, not, “Can I help them myself?” I think you see that very much with things such as the six-point plan and so on.

DWP can tell you about the other points that they have. We have not looked at them in detail but there are things such as the ACSLs—the advanced customer care service leaders. Sorry, I think I got the acronym wrong but you are nodding so I think you know what I mean. You have that kind of role. I think you also have something about—

Chair: Advanced customer support senior leaders.

Joshua Reddaway: You can tell that I have not done an audit on this but the accounts refer to them as being quite key.

The second kind of relationship to vulnerability is with those who are specifically vulnerable if their benefits are stopped. They are vulnerable to decisions made by the DWP. The DWP has a role but it is one as set out in the regulations. If I may say so, I think that Mr Mills asked the question quite well at the last session about the tension between what rules are set and how you consider harm but the rules do have things such as the safeguards in PIP assessments for assessing whether or not people are at risk of harm and that acts as an override to the points system. That is something that Parliament needs to be interested in—what rules are being set and what safeguards are in place.

The third way that DWP thinks about vulnerability is the one that we have mostly focused on in our work on DWP and that DWP has set out in its written evidence to you. That is some sort of vulnerability, however defined, that prevents people from accessing DWP’s services. Of those things, the common ones that we saw were digital literacy, language skills, being socially isolated or the habitual residency test and, therefore, where people are marginal in society.

For example, we looked at payment timeliness and the Department had done a very good of getting payment timeliness up from around 55% of UC claims being paid on time to around 90% and it has been hovering around 90%, or is just below that at the moment. But what could they do about that last 10%? The Department said, “It is not us, it is the claimants” so we looked at that last 10% and in every one of the cases we looked at, there was something more complicated going on. It was not just the claimant saying, “No, I refuse” and I think we would call that a vulnerability to not being able to access services properly. That is the way we look at it.

Chair: Thank you. Debbie Abrahams has a quick point.

Debbie Abrahams: Joshua has probably already answered my point, which was about us all as human beings being potentially vulnerable to different harms and whether that was recognised. I think you are saying, Joshua, that that is the case and those three definitions that we should be considering reflect that. It is not just about identifying vulnerable claimants. It is about recognising as human beings that we have the potential to be vulnerable if we do not get our money in time, if we are not treated with dignity and respect and so on. It is those different aspects that we need to consider. Thank you.

Q63            Sir Desmond Swayne: The Department has something called ACSSLs—advanced customer support senior leaders—and one of their key deliverables is, “Building and maintaining relationships with external organisations that support vulnerable citizens who may also be customers of DWP”. Are they any good at it?

Joshua Reddaway: That is the acronym that I managed to bungle in answer to Sir Stephen’s question in saying that I have not looked at the role in any detail. However, I think that kind of system that you have in place is very important in supporting work coaches.

What I have looked at is how work coaches support people who come into the job centres. The most relevant work was on how they support disabled people through the intensive work search where they are supported by disability advisers and a network of others. My assumption is that these advanced case leaders are essentially another layer and according to the accounts and the definitions, they are plugged into wider networks. I hope that you will be able to get a lot more from the Department than you can from me. The ability of a work coach to do it alone is very limited, so they are very dependent on the support networks within the Department.

Q64            Sir Desmond Swayne: Ashley, two reports, two reviews, have accused the multi-agency safeguarding arrangements of being disjointed, fragmented and not yet fit for purpose. How can that be fixed?

Ashley McDougall: One thing is time. These were new organisations so there has been some understandable feeling of their way as to who has what data and who has what responsibilities. Sometimes the problem is personal—the actual players from local organisations who will relate. I think what will work are clear ground rules for what they should be doing, what data they need to be looking at, what they can do with it, what the responsibilities of the individual organisations are and how they can relate to other organisations. Otherwise, a bad outcome is a talking shop that people will look back on afterwards and say, “We didn’t really know the full picture”.

They need the data to understand who is vulnerable, where the trends are and what they can do about them. They then need to have actions—basic governance, minuted actions, follow-up and outcomes, whether they are improving the outcomes of the local area. That is how they would need to move. One of the two reports was done by the chap who recommended that they be set up in the first place, Sir Alan Wood. They say they are just not where they need to be and that is a combination of all of those personal, organisational and data-led problems.

Q65            Sir Desmond Swayne: There is a programme that the Department for Education, the Ministry of Justice and other Departments have been involved in—I think it is called BOLD—intended specifically to deal with fragmented data sources. Could the DWP have picked up something from that? Should the DWP have been involved in it?

Ashley McDougall: The Department would not have been involved in the exercise the MoJ and DfE did, which was looking at school outcomes for people who had been involved with the criminal justice system, people who had been suspended for instance. Many of those who had been suspended from school had ended up with sentences or had come into contact with the criminal justice system in some way. There was probably no role for DWP in there.

One of the things we talk about in the report is NEET—children who are not in education, employment or training. There is a data analysis job to be done on that population. How many of that population—who are in no way criminal, but you could apply the same techniques and expand the analysis. This is really about the pipeline—to ask how soon we can identify people who are at risk of falling into bad outcomes by joining datasets and doing something about it. NEET was one of the outcomes we looked at and that approach seems to me to be a perfectly reasonable extension of that data.

Q66            Sir Desmond Swayne: The National Audit Office has told the Department that it needs to develop a better data-based understanding of the numbers of vulnerable claimants and use this to support the needs of people who continue to struggle with making a claim for Universal Credit. One of the criticisms was that it did not use data flags to mark those vulnerabilities. Has it got any better?

Joshua Reddaway: Yes, a little. The problem we identified had to do with the “being vulnerable to accessing the services” definition of vulnerability and the Department’s ability to target those people who require more help and to identify them as soon as possible, for example not waiting for the payment to be delayed before providing help. The main help for Universal Credit, of course, is the service provided by Citizens Advice and at the moment there is no way of triaging claimants and referring them to that service, or of using data to do so.

The bit of the system that the Department changed recently—by which I mean a year ago—was putting in some voluntary questions at the beginning of the claim to at least ask people what the barriers to making a claim were for them. The promise was that the Department would look to see whether or not that data could be used by work coaches to provide better support during the claims process down the line to better triage and evaluate what was going on. I saw this as a baby step in the right direction.

There are a host of data issues around vulnerability, which we have raised previously with this Committee and with the Public Accounts Committee, but that recommendation referred to a very specific issue with the Universal Credit system.

Q67            Sir Desmond Swayne: If the aim—and it is the aim of the Department—is to provide a more tailored experience to claimants, how does it get that understanding of what the claimants’ vulnerabilities and needs are? How is going to fix that?

Joshua Reddaway: I think one of the biggest challenges the Department faces over the next 10 to 20 years will be how it is going to provide modern, digitalised services. I say that sort of timeframe because we know from the implementation of Universal Credit, and also the work we have done on the health transformation programme, that you are talking about 10-year programmes to digitalise things.

The fundamental issue is that the benefit system is a rules-based system that has legal decisions, a universal process and universal criteria, but you want to tailor it to the needs of individuals so that they can better navigate through it. That is very much what the Department has been trying to do with the child maintenance service. I gave evidence to this Committee a year ago on that and the way that programme was going. It is what the it wants to do with the health transformation programme, which is essentially that when you apply for PIP there will be a triaging at the beginning that will dictate the process and the form of decisions. They are the same criteria but how those decisions are made will depend on the triage, the idea being that certain conditions will not need to go through the same sort of assessment process as others—you will not need visits or to see someone face to face and so on. That should reduce costs and give people a better experience. Universal Credit predates some of the ambition for tailoring; the process that people go through is still very much universal without much tailoring.

That is the grand scheme of things and we are talking about a big, big challenge for the Department. I can go into detail on the individual bits of data but I will let you ask if you want me to.

Q68            Dr Ben Spencer: We have spoken before about tailoring versus generic thinking, the fundamental tension between having a system that works and works quickly versus using more qualitative data and being able to add a bespoke experience.

My question relates to some evidence about multi-agency safeguarding and evaluating the effectiveness of those systems and it is an open question. If you are looking at models of multi-agency working, how do you go about evaluating effectiveness if you are going to compare different models, structures or ICSs across the country? How can you rate them and compare them? What sort of datasets and metrics do you look at?

Ashley McDougall: We want to make sure they had a focus on outcomes and whether they were changing things. The process we go through is, at the basic level, do they have the right people in the room—the NHS, the local authority, the police—and people at the right level. It might be only junior officials but it would need to be decision makers who are aware of the risks. Escalating from that, you look at things such as the number of children taken into care. That is one of the things we want to know. We want to have the right number taken into care and the data that will allow us to say whether we are taking more into care than parallel areas. There will be comparisons with other partnerships. You don’t want to compare rural with urban; you want to compare like with like. Once you have the data, you need to compare it to admissions.

We had a sectoral report on access to mental health services, one of your areas. How does it compare from ICB to ICB and how does it relate to other aspects such as contact with the criminal justice system or absence from school—schools being one member of the partnerships in the multi-agency safeguarding? That is an outcome that you want to track. Things can go wrong, and multi-agency partnerships are supposed to be joining up those services to make sure that fewer things go wrong, looking at trends and making comparisons. That work has to be data-led because otherwise you will just have a lot of talking, rationalising and narratives. You need to have the data and you need to be clear about the outcomes you want to achieve through the partnerships.

Q69            Dr Ben Spencer: A lot of that is procedural evaluation, making sure that if we say this is how the system should work, it does what it says on the tin, is working how you think it should be working. I am curious about the number of possible confounds in doing that sort of comparative work on outcomes. Some of these outcomes are more conceptual in the sense of being outcomes that you want to avoid but nevertheless are outcomes that will be necessary. How do you go about evaluating that? A child being taken into care is an awful thing but it is not awful if the child needs to go into care. How will you go about evaluating the system in those terms? That is what I am trying to understand.

Ashley McDougall: It comes down to the comparators. Say you have a similar population—we talk about model authorities and model areas—but their trends have been going in different directions from yours, first you would want to know if you are an outlier for the number of children taken into care. That is why I hedged earlier—you do not want to get that number down because you want it to be the right number, but what the right number is will probably be well informed by comparisons with other authorities with similar populations. You find your partners, do the statistical comparisons, see which way you are going and that is the way you would do it.

I think there is a question as to who will be doing the statistical evaluation. Alan Wood, who set them up, and the child safeguarding national review panel, who came and looked after a couple of terrible child deaths, looked at it episodically. There is no one looking at it systematically to see how well those partnerships are doing. How they are working is up to the partnerships but there is no systematic regular evaluation and that would be down to the Department for Education.

Q70            Dr Ben Spencer: Just so I have this clear, there are no models in operation in the UK where DWP has a seat at the table. The DWP is pretty much excluded or is not a party to any of this.

Ashley McDougall: That is correct.

Q71            Chair: Can I ask a very quick question? A couple of years ago, the then Secretary of State came and talked to us about what the Department was doing on safeguarding and talked about a network. Was that the point when these advanced customer support senior leaders were introduced or is this longer standing?

Joshua Reddaway: I was trying to work that out in preparing for this Committee. That is my understanding. I would prefer for you not to put a huge amount of weight on that and to ask the Department, but that was my understanding.

Chair: They are fairly new?

Joshua Reddaway: For example, I think Peter Schofield told you that they would be doing case conferencing calls where necessary. My understanding is that this is the response to that but I have not done any audit work on that. That is just from my reading around it.

Q72            Selaine Saxby: Joshua, this question is for you. I know you have already covered a lot of this but I want to focus specifically on the accessibility of benefits and the main challenges facing vulnerable claimants when they make a new claim.

Joshua Reddaway: My main audit evidence for that is from when we were looking at Universal Credit and getting to first payment. We were interested in the 10% who are not paid on time. What generally happens is that the system will be saying it is still waiting for some information from this individual and we would ask why—the claimants presumably want to get the benefit. We looked at that and tried to work out what the barriers were and it was really clear, just from following the cases, the history and the communications, that we were talking about people whose language skills were not great, whose ability to understand instructions, either because of language or for some other reason, was not great, so that when they were being asked for things, they did not seem to be responding in a way that showed that they knew what to reply.

After talking to lots of the stakeholders—and of course you have a panel of people later on who have spoken with the benefit claimants much more closely than we will have done—we understood it be things such as lack of digital literacy, language skills and the habitual residency test came up a lot. Whether or not you want to call that a vulnerability, we wondered if it was correlated with people who were ethnic minorities from other countries. We wondered if that was associated with that, but it was also because the immigration system and habitual residency test is a minefield of its own. That, plus these other things, was what we saw.

Q73            Selaine Saxby: Given that, you said a lot of it stemmed around language and understanding. Is there more that you think that DWP could do at the time of application to avoid it getting stuck in the first place? Are there things that you can identify in the process?

Joshua Reddaway: We said it would be really good if they were able to triage these and work out who needed the additional support. Of course, Citizens Advice are the ones that are being paid to provide the support for Universal Credit. We also said that that might need to be competed and put on a more official basis, because it is essentially a grant.

The Department thought about it and decided not to because, apart from anything else, I think it wants to protect the independence of Citizens Advice, which is reasonable. Where we were really coming from was the fact that at the moment it is a very standoffish relationship. They don’t get much information from Citizens Advice, so they are not really doing this in a data-led way. They don’t really know who they are referring. Does that support provide the right support and does it actually help? Do people who get referred to Citizens Advice have an easier time to go through and get paid on time? It is basically like a blockable grant. You throw it over and you don’t really know what happens. We would like to see it put into a management information and data-led process here.

That goes back to the fact that I think DWP has an awful lot to learn if it is going to come against the challenge of tailoring its services to individuals rather than offering universal systems, processes and things that work for—well, in this case it literally works for 90% of claimants and not for 10%.

Selaine Saxby: That brings us back to Ben’s question. Thank you.

Q74            Debbie Abrahams: Good morning again. My questions are for Josh and are particularly about the report that you produced in 2020, on the information held by the Department for Work and Pensions on deaths by suicide of benefit claimants. So that everybody understands, this was commissioned—and do correct me if I get anything wrong here, Joshua—by the former Work and Pensions Select Committee Chair, Frank Field, after he asked questions to the Department that could not be answered and basically you were to investigate deaths by suicide alone of claimants.

Again, please correct me if I get this wrong, the key points are around 2016-17. There was a coroner’s focal point established by the DWP; is that right? Within that, 19 contacts about deaths were made from coroners, of which nine were suicides. However, not all coroners had been using it and, as I understand it, are still not using it, so it is a bit of a hit and miss process.

You then looked at the suicides in the IPR process and you identified 69 suicides just for the year 2014-15. You also identified that in 2013 there had been four section 28—the prevention of future deaths—notices to the Department for Work and Pensions, but you identified that it is highly unlikely that all cases that could have been investigated by DWP have been investigated. Having gone through that process, do you have a view of the likely order of deaths and, within that, the deaths by suicide by claimants?

Joshua Reddaway: You did ask me to correct you so—

Debbie Abrahams: Yes, please do.

Joshua Reddaway: That is broadly correct. We were not commissioned as such. We were asked—basically, Frank Field said that he had been quoted that it would cost too much to get this information. That seemed to us to be an audit question, so we looked at the cost. It turned out that there was some information, but it was not complete information and that is what we said.

You have quoted the correct numbers, but those are since those dates. Therefore, the way to interpret it would be 19 since March 2016 to the time we were appointed in 2020, for each of the others. They are over multiple years and unfortunately over different years, so they are not easily comparable but it is a full list of what we had.

Since then, I have not done any specific work on this. We have the new disclosures that were in the accounts last year. I was very pleased that the Department chose to put more in place. That appears to have been a very active decision by the Serious Case Review Panel. You can see in its minutes that it decided to do that in the March of that year. It wanted to better explain what it was doing.

I note that you have picked up in your inquiry remit and request for evidence that the number of IPRs seems to have gone up a lot—

Q75            Debbie Abrahams: Thanks for that. I have questions on the IPR process, but this is specifically about identifying deaths. What I think you are saying, Joshua, is that you can’t really say whether that was in the order of deaths—and I appreciate what you are seeing in the timespan, and we have seen a trebling in the IPR reported deaths, but in identifying deaths—

Joshua Reddaway: I have no new information other than what is—I am completely reliant on what we put in that report, what is in the accounts and what you have seen.

Q76            Debbie Abrahams: Basically, we don’t know the order of deaths coming through the coroner process, which of course will not be exclusive. We know that it is only unexplained or unexpected deaths that will go to an inquest process and there may be others as well.

Are you aware that in 2019 there was a Freedom of Information request to the DWP that revealed that 274 claimants a month who had been found fit for work subsequently died within six months? The same year there was a written question by former MP Madeleine Moon on claimants refused PIP. The response to that question was 60 claimants died in three months, having had their PIP claim refused. This was not within your remits, so you may feel that you are not able to comment on it, but do you think that we are doing enough to understand the cost of human life in the deaths of claimants?

Joshua Reddaway: I am not familiar with that particular FOI or those statistics. What do I think? My immediate question is: died from what? On the destination of what happens to people after benefit decisions, I am afraid I am going to take you away from what you are interested in, but I think that is an underexplored area. The statistics are not very good at showing what happens after, for example, you make a decision that somebody is fit for work—do they actually come in and claim UC and do they then find work? It was one of the questions we have never been able to understand, including when we were doing our work in 2019 on support for disabled people.

Whether or not they died is an easier statistical question because I can imagine that the data is there, but it does not necessarily mean a lot because there are always people—

Debbie Abrahams: There will be some deaths, yes.

Joshua Reddaway: Sorry, it is not a very useful answer, but that general lack of data is the kind of sense.

Q77            Debbie Abrahams: You could compare them to the working age population as a whole to get an understanding around that. How do you think the Government should be monitoring and investigating harms to and deaths of claimants?

Joshua Reddaway: I was very pleased to see that, like I say, the Department put more emphasis on this after we did our review. I think one of the reasons that there are more IPRs now than there were then is because—I would not necessarily interpret that as there is more of a problem because it may or may not be that there is more of a problem going on with cost of living and so on. It is definitely, at least in part, because they are putting more attention and more focus on this, and they have set up a team and they are showing the care to do that.

I think that what you would expect to see is a learning process within the Department with actions. The questions I would have—and I have not audited this—and questions that you might want to consider are: do they have the follow through? The accounts show the number of actions that are closed, and I have seen that has been FOI-ed—the list of what those actions are—but do you get a sense of whether or not that is leading to the right sort of corrections? I don’t know; I haven’t audited it.

Secondly, do you think there is the right level of transparency in the system? I think some of this is very private and it needs to be if you want them to do the proper lessons, learning and to improve things. Equally, I find it very interesting that there is a system in place for complaints that ultimately has a system of independence if you can go through your MP to the Parliamentary and Health Service Ombudsman, the PHSO. That is not available for investigating serious cases if the person has died, because my understanding is you have to be alive to make a complaint.

I think that would be a question: is this system dependent on the current culture and the current interest of the Department or is this a system that is sustainable and has that level of accountability over it? I think that is a really reasonable question.

Q78            Debbie Abrahams: Thank you. That is very helpful, Josh. Thank you very much indeed. It is only investigating case mishandling for IPRs. It is looking at what at an operational level has gone wrong. Do you think there is an issue around where safeguarding actually begins? Should the safeguarding approach start at a policy level and follow through right to the operational end, so that we understand what could potentially be creating vulnerability, as we discussed earlier?

Joshua Reddaway: I think I could be getting very close to commenting on policy, which I am statutorily not allowed to do in this role, if I was to say what the policy is and how the policymaking should work.

Q79            Debbie Abrahams: Ashley, do you want to come in? Where should a safeguarding policy start? Is it just about the individual who is delivering the particular service one to one or should it start at a policy development level?

Ashley McDougall: It is one of the aspects that you have to have in policy development because you need to identify the risks, the benefits, the trade-offs that you need to apply. When the Government do—there is a name for them that escapes me but essentially for policies that have cost benefit impact assessments—those impact assessments, they would really want to know all the dimensions that might be affected and what metrics they might be looking for. I would expect that to be part of what they should be doing.

Q80            Debbie Abrahams: I will press you about what sort of metrics we should be looking at, both at the IPR and you might want to comment on other stages of safeguarding.

Joshua Reddaway: What do I think? It is really difficult. You can have specific management information that is about the process, which is what they are trying to give you by telling you about the IPR and so on. That gives you some sense of assurance that they have a system in place. It does not really tell you: is that system working? Ultimately you need to have good management information, and preferably transparency over that, about customer service and customer outcomes. They do have good transparency around payment timeliness and that is why I am able to tell you why there is 10% and what is going on with that 10%. However, do they have that across the wide range of things? I do not think so. We are currently looking at that, so I should reserve judgment slightly given the study that my colleague Laura Brackwell and the team who have taken over from me are currently doing.

I will give you an example. We were looking at the way they use AI in the detection of fraud. The Department has not quite done what it said it would do yet, but it has promised to properly assess those models for bias. That is not about vulnerability per se but it is about bias towards certain protected characteristics and that could be ethnic groups or whatever, so it could be related vulnerability. It was kind of promising a very technical answer about what a model says that did not actually tell you what the outcome was. We have recommended that if the Department had better customer service information and could tell you what the impact is on people of introducing that model that would give you more assurance. Do you understand what I am trying to—

Debbie Abrahams: Yes.

Joshua Reddaway: I am worried that I am not being clear. There is that distinction between outputs and outcomes. If you are not told what the outcomes are for people and what the exceptions are and zero in on the exception to the outcome, you cannot really say how you are treating vulnerable people. You are talking about a very small number of the exceptional cases, I think, out of the millions of people that DWP deals with, a percentage to 90%—it does take in about 10%--but if you are after the instances of the five out of a million you need some sort of regime that will measure that and give it to you.

Debbie Abrahams: Yes. Thank you so much.

Q81            Nigel Mills: Joshua, in your work did you feel that DWP was taking seriously the findings of the IPRs or various other complaints processes and actually changing things or did the complaint processes just happen and life carried on as normal?

Joshua Reddaway: In all honesty, I don’t think it was taken that seriously prior to our review and I was very interested in how—there is a paragraph at the end of our report that sets out what the Department wanted to do. That was just to open the curtain on how our reports are produced. That was not our suggestion. That was very much the Department saying, “We are happy with this report, but we have to set out what we are going to do about it”. That is what it asked to add and we were like, “Well, it is not audit evidence, but it just says quite clearly the Department says it will do the following”. I think it did go on to do those things, as far as I can tell, and it has been reporting on those things.

Therefore, I think there has been a little bit of a step change in attention on this by providing some—which is obviously what we want through the accountability process and scrutiny, isn’t it, to drive a little bit of prioritisation.

Q82            Nigel Mills: The Secretary of State wrote to us last week on the Department’s use of AI and suggested that one of the uses was to be able to scan correspondence and spot vulnerability earlier or more effectively than humans could do it. Is that something that you have come across it doing?

Joshua Reddaway: I have not looked at it in any detail, but I am aware that it is doing that, and I think we will be including it in a case study in the future, a piece of work we are doing across government, AI or something very similar or some of the DWP stuff there.

From looking at the implementation of AI in counter-fraud, it definitely has merit in rollout in that sort of area. From the large language model stuff that we are seeing, the ability to detect trends and patterns and be able to spot stuff, given the very large volume of stuff that DWP is dealing with, I think it is a particularly fruitful area.

Q83            Nigel Mills: It is kind of like the reverse, isn’t it? When you are looking for trends you think, “I have suddenly got 1,000 identically character-for-character applications. That is very odd. That must be an exception.” Here you are looking for, “I have a very unusual piece of correspondence that should set alarm bells ringing”. I guess that is a bit harder to ask a computer to spot.

Joshua Reddaway: My understanding is that DWP is looking for the application of AI to support human review. For example, when you have a big mailbag of correspondence and you are triaging in what order you deal with that, AI might be able to support that triage process, but it does not stop the need for a human to read it and go, “This one wasn’t flagged, but something is odd here”.

Ashley McDougall: We have looked at all sorts of systems and you would have trigger words, so it is not just the trends for the big volume that AI might be able to help you with. It is scanning big volumes for the particular things that you are looking at and you could use that to focus it on the micro. The National Audit Office has some systems for our internal feedback systems. They will look for words that the office wants to see. Is there anything that talks about bullying? If anyone mentions that in any of their comments, it will highlight it and use that as feedback, so the systems can look at the very big and highlight individual things.

Q84            Nigel Mills: Joshua, in your years of auditing DWP, I sense you qualified every set of accounts, I think on regularity, wasn’t it?

Joshua Reddaway: It was 1988.

Nigel Mills: I think from the roundtables we have had and the other evidence, what some vulnerable people want is to say, “I am so vulnerable, I can’t cope with the reassessment” or “I am so vulnerable that refusing me my benefits will be so harmful that you can’t refuse me my benefits”. Are those options where DWP has to have discretion and say, “This person would not normally meet the rules” or “They should normally be reassessed but we think that the harm could be so great that we will effectively give them benefits that technically we don’t think they are entitled to”? Is that within the Department’s power or does that leave it in the regularity thing that the Parliament has set the rules and it has to follow them or not diverge too far from them?

Joshua Reddaway: This is what I was saying about vulnerability that makes people particularly susceptible to harm as a result of the benefit decisions, and that there are certain safeguards that are built into the rules. However, that is very much at the risk of me getting into the policy space because I think these are the rules that are set by Parliament through the regulations, and it is for you to consider what the safeguards are.

A good example is the one for personal independence payment and the substantial risk category. This acts as a non-functional descriptor. Essentially, you can do the points-based system and then you have this, “Okay, notwithstanding all of those points and what they have actually scored, is there a substantial risk and how is that assessed?”

I have not had the pleasure of sitting in that many PIP assessments, but this came up in the first one I sat through watching. It was at the time when they had tightened the guidance, so somebody who had claimed to be a suicide risk was being determined not to be a suicide risk any more, and unfortunately that meant that her benefits were very likely to be cut. I could see the kind of emotional difficulty and challenge that was for this person.

My expectation is that Parliament thinks very carefully about these safeguards and what it wants to get out of them. The question Parliament might want to have for the Department is real clarity about what is the impact and how you judge the effect, where to get to the line on these safeguards. If they are too generous, that is a bad thing, presumably, because it reverses the point of having the eligibility criteria in the first place. If they are too tight, there is a risk to people. I am not convinced that we have the management information to allow us to judge that or that you have that information, and I would be interested to see what the Department says about that.

Q85            Chair: Thank you. Josh, you made the point that part of the IPR process necessarily is rather secretive, but you also welcome the fact that there is some new information in the annual report. It does look quite an opaque process from the outside. What do you think the next step is in opening up a bit a bit further?

Joshua Reddaway: I think it is being clear about the disclosures that the Department will be making. It has changed every single year and the bit of the annual report and accounts that it is in is the performance report. Therefore, I need to give you the clear warning on that, that is the unaudited part of the ARA. We have a responsibility to make sure that—we call it a consistency check—it is consistent with our understanding, so everything in there. I would say particularly the fraud and error bits, because we have a lot of understanding of that. However, in this case it was consistent with our understanding, but it is not audited and reviewed.

There is a lot of leeway for the Department to choose what it is reporting, so it is quite possible that this year’s disclosures on this will not be in next year’s or in the year after. I think it is being clear about what you want to see, do you think this is the right mechanism to get that out and what are you expecting to see trend data on, how things are changing overtime. I think you could help the Department by setting out what your expectations are.

Chair: Thank you both very much indeed. That concludes the questions we have for you. We are very grateful to you both.

Examination of witnesses

Witnesses: Brian Dow, Daphne Hall and Jamie Thunder.

Q86            Chair: Welcome, all of you. Thank you very much for joining us. Let me start, as I did with the first panel, by asking each of you very briefly to tell us who you are, starting with Brian Dow.

Brian Dow: Yes, lovely to be here. I am Brian Dow. I am the Deputy Chief Executive of Rethink Mental Illness.

Daphne Hall: I am Daphne Hall. I am the Vice-chair of the National Association of Welfare Rights Advisors. I also work at Rightsnet, which is a social welfare law website.

Jamie Thunder: Good morning. I am Jamie Thunder. I am the Policy and Public Affairs Officer at Z2K, which is an anti-poverty charity that also provides a range of advice, support and representation around social security and housing.

Chair: Thank you all very much for being with us this morning. The first question is from Dr Ben Spencer.

Q87            Dr Ben Spencer: My question is, focusing on the implications and implementation of a statutory safeguarding duty, what would a statutory safeguarding duty achieve over and above good or improved safeguarding procedures?

Daphne Hall: I don’t mind starting. I think it brings in a level of accountability and enables oversight. It goes back to what Joshua said in his last point, that certain things have to be disclosed and then we can look at them and see how the DWP is performing overtime. I think it gives that accountability particularly.

I like what was being talked about before, the local things that they may be using at local levels and involving he Department for Work and Pensions. I think it was said it was NHS, police and local authorities, but DWP should be that because sometimes local things can really help in safeguarding local knowledge.

Brian Dow: I completely agree. Implementing a safeguarding duty would ensure that the DWP is more effective at performing or walking that balance between protection of the people who are most vulnerable and will require long-term support from the benefit system and graduating those back into work who may be able to at some point in the future or who are currently fit for work. Fundamentally that is because there should be an accompaniment piece between an organisation’s safeguarding duty and the culture that it has to learn lessons and develop and enhance its performance.

Jamie Thunder: I do not think that there is any particular reason why in principle DWP could not implement the same sort of safeguards that you might envisage under a specific duty without one, but it would have the effect, as Brian and Daphne have said, of raising the salience of that with the organisation and giving something concrete to hold the Department to any such duty, as well as looking at the crystallisation of risks in instances such as death or very serious harm. It also needs to look at the risks that are being run within the processes. There is a lot of ways where even if a risk does not crystallise in a serious event, there has still been a very high level of risk that someone has been exposed to.

Daphne Hall: Could I just add one thing? In policy development it could be part of every impact assessment when new policy regulations are coming in. If safeguarding was there as something that had to be addressed in the impact assessment, hopefully it could catch things that might go wrong early on before the law even comes in.

Q88            Dr Ben Spencer: All of you have mentioned accountability in some way, shape or form. Who is accountable?

Daphne Hall: At the end of the day, it is at the top; the permanent secretary and Secretary of State would have the ultimate accountability. I am not an expert on safeguarding but I looked at the CQC website and it seems to say that the ultimate accountability is at the top level. However, obviously within that is training at every level because it starts at the bottom. It is the frontline that is absolutely key, but the ultimate accountability is at the top.

Q89            Dr Ben Spencer: Would work coaches be held individually accountable?

Daphne Hall: No, not in my view. They should be trained and supported and have a clear line where they can report concerns and stuff like that. If they are struggling with it, they should be offered support. However, no, not individually accountable. Something has gone wrong if a work coach has not had the ability to feed stuff back up. There is a lot about resources, which I might come back to later on.

Q90            Dr Ben Spencer: For me, this is so important and key to understand and get right. If something goes wrong and a work coach is found to be at fault, you would exclude them from their fault under the statutory duty of safeguarding?

Daphne Hall: Perhaps not for that. Obviously, we would need to look at their work and it would be a management issue. However, at the end of the day, something has gone wrong with management that that has not been picked up.

Q91            Dr Ben Spencer: What about managers of jobcentres? Are they within the scope of the statutory duty statutory?

Brian Dow: I am not sure about the answer to that, but I agree with Daphne that what you have to have in any system of accountability is a tier. Clearly frontline staff have to be trained to provide the correct service for people. There is a training and compliance dimension. This is true in the education system, it is true in the health system and it is true elsewhere. Management has to take responsibility to enhance and improve performance, but that goes right the way up to the top of the organisation. Quite where you draw that line I would not want to place a bet on right now.

Q92            Dr Ben Spencer: Is there any model that we have in which individuals are subject to a statutory safeguarding duty through their work in which they are not professionally regulated, or regulated in some form? The examples that I can think of are people who have statutory duties—nurses, doctors, social care workers. They all have some degree of professional qualifications or professional regulation. I am certainly not an expert in this area. Are there any other areas where people have statutory responsibilities for safeguarding but are not professionally regulated or regulated in some form? Is this a novel thing that we are contemplating?

Brian Dow: I don’t know the answer to that. I couldn’t tell you.

Daphne Hall: I don’t know the answer, but when you are looking at it certain things need to have safeguarding, certain decisions, so it is a bit more focused. If you are going to make a decision to withdraw benefit, that is subject to certain things and that would be different. Not just an all-sweeping statutory safeguarding but focus it a bit more and then maybe it is easier to regulate and see who is responsible.

Q93            Dr Ben Spencer: For me this is a fundamental question, which is about a duty that requires people to behave in a certain way. What is the scope of that for accountability and responsibility? It seems from what you are saying that you see it as a statutory duty for the Secretary of State and the permanent secretary but not for anyone else.

Brian Dow: I think that if you have a statutory duty it covers the system as a whole. I am not a legal expert, but the responsibility for how you apply that is clearly different depending on where people are sitting within that system. You would ultimately expect the Secretary of State to take accountability, as the Secretary of State for Health does if there is a serious incident in the health system. However, you would not hold the Secretary of State responsible for a failing in an acute hospital, for example. Ultimately it goes all the way up to the top.

Q94            Dr Ben Spencer: In an acute hospital you have governance risks. You have the CEO who has responsibility for the hospital functioning, but the people working in the hospital are all regulated, the majority of whom are regulated in their duties. It swings both ways in that sense.

Daphne Hall: One of the difficulties, particularly for work coaches, is the level of pressure that they are under. You are probably aware that the PCS Union handed in a dossier of evidence to Caxton House last week of testimonies from work coaches. They are quite horrific. They are under such enormous pressure that they do not have time. It is a management or higher-up issue that the resources are not being put in there. That makes it even harder to hold them to account when they do not have the resources and the time with the claimant.

Q95            Dr Ben Spencer: Wait a second, if they are responsible for a statutory duty, I can’t see how they will not be held to account in some way. Then you are saying that they cannot deliver it. Are we not creating an impossible situation?

Daphne Hall: That is why I was saying perhaps to focus it on certain types of decisions, like decisions to remove benefit. The work coaches do not make those decisions, that is higher up. Maybe that would be a starting point where you focus on specific decisions that are likely to cause harm, rather than day-to-day, every discussion with the work coach. Those decisions should have a safeguarding duty and there would be a much clearer line of accountability, and only a certain level of staff could make those decisions.

Q96            Dr Ben Spencer: Do work coaches want this?

Daphne Hall: Do work coaches want to have safeguarding duty?

Dr Ben Spencer: The statutory duty.

Daphne Hall: I do not think that it could be put on work coaches. I would be very surprised if they would want it, given the pressures that they already have on them. I would be very surprised.

Q97            Dr Ben Spencer: I was interested to read, Brian, the list that Rethink Mental Illness put together of suggestions of what a statutory duty could involve. With the exception of (f), which is the appointment of an independent regulatory body to enforce a statutory duty, it strikes me that (a) through (e) and (g) can be provided without a statutory duty.

Brian Dow: That is possibly the point. I should have bought our submission along to have that in front of me. I suppose, if I can slightly dodge that question, because I do not have it in front of me, and make an associated point, your line of questioning before is a good one, about where you apply that duty and who is responsible for it. I did a search before I came in here about the visibility of the statutory duty across DWP and I compared it to the NHS. After about 10 minutes looking on the DWP website, I gave up. It took me all of five seconds to discover exactly where it applied in the health system—you would be very familiar with that, of course—and precisely what that meant in practice.

There is a germane point about how you apply a statutory duty, who is responsible and how you do not make the system worse and put greater pressures on people. However, if you compare the architecture that exists around the Department for Work and Pensions to the other big areas of state, there is a very sharp contrast that is hard to deny.

Q98            David Linden: According to the Money and Mental Health Policy Institute, a recent FOI request showed that 38% of customer-facing staff in DWP had not had mental health training. Mr Dow, can you think of any other apparatus of the state that requires adults to interact with it that would not have that kind of support level in place?

Brian Dow: No, not really. There has been a long-standing debate about how well trained people in the benefits system are around mental health. By way of preparation this morning I had a look at the WCA form and the Universal Credit form and it is pretty clear that both in the form and the assessment process itself, there is not a good enough understanding of mental health conditions, exactly what a mental health condition is, the extent to which it is often a very fluctuating condition for somebody, and how it could be that somebody with even quite a severe and enduring mental illness is able to work.

One of the conclusions that I drew in reacquainting myself with it is, to be frank, how somewhat patronising the process is. Over the years that I have worked at Rethink I have met engineers, bankers, architects, politicians, dare I say, who once would not have been able to do their role, who have had mental illness. I think that the benefit system as a whole—I should say, by the way, that I spoke yesterday to very senior officials in the DWP and I feel very confident that there is a wind of change moving through the Department. However, there is simply not a good enough understanding. Therefore, no, I could not tell you another area of public life where that is the case.

Q99            David Linden: Is it a fair assessment to say, Mr Dow, that as a society we have come some way in being able to talk about mental health, understand the importance of treating it in the health service, but is it also fair to characterise it to say that DWP policy has not necessarily caught up with where society is at?

Brian Dow: I think so; I agree completely. There is a fundamental problem, of course. The DWP has to carry out a very difficult responsibility and walk a very difficult line between, on the one hand, ensuring those who require support receive that support. The burden of mental illness has grown over the last four or five years and I am sure that members of the Committee will have seen, for example, very worrying figures on the rise in mental ill-health among young people, which is terrifying. It has to make sure that it preserves the rights of those individuals and that when they become adults, it helps people into work that they are able to do. The nature of work, of course, has changed as a result of the pandemic as well.

It is true to say that to some extent the public are ahead, historically at least, of where the DWP has been. Its machinery is not particularly good. If you look at the assessment process as a whole, the forms themselves, it is pretty clear to me that mental health plays second fiddle to physical health, and the interaction between the two, which is circular, does not seem to be very well understood. There is a number of quite serious cases, including of people who have died as a result of that misunderstanding.

Q100       David Linden: My colleague from Oldham East and Saddleworth has made reference to them on many occasions in Parliament. If we take Ms Hall’s point that staff should be trained and supported adequately—I am familiar with the document that the PCS handed into Caxton House last week—you will be able to see in the line of questioning that there is a difference of opinion on the Committee about the role of introducing a statutory duty. If we park that to one side, aside from introducing that statutory duty, what other changes are needed to ensure that DWP takes appropriate responsibility for safeguarding those vulnerable claimants?

Daphne Hall: We had a conference last week with the trauma lead for DWP, who is fairly new in post, on the whole trauma-informed approach. There is something on gov.uk about the trauma-informed approach. There are the six pillars, which are safety, trustworthiness, choice, collaboration, empowerment and cultural consideration. That is a good starting point if everybody in that trauma-informed approach does not assume things when you first see somebody; not, “These are the things you’ve got to do”, but “What do you need?” That should be the starting point.

That goes back to the previous session on where do you start. That is the start. When somebody fails to interact, that should be the first question, “What do you need?” Not, “This is the long list of to-dos that you have to do; if you don’t do them, we’re going to close your claim, even though we shouldn’t.” “What do you need?” It is asking somebody and taking them along with you, that whole empowerment, giving them choice along the way. It does not have to be digital; if digital doesn’t work for you, that is fine, we’ve got another way of doing it. That trauma-informed approach is a really good starting point.

I was very encouraged by Bea, the woman who is doing it, and I know that they have Lisa Scullion for Salford University doing some research, who is absolutely excellent. That is a brilliant starting point. Every single person, particularly the frontline person, should be trained in a trauma-informed approach and how to deal with people in a trauma-informed way. I think that that is a great starting point.

Q101       David Linden: Do you think that work coaches with their existing workloads have enough time to get trained?

Daphne Hall: They haven’t got a hope. All these reforms that the DWP is bringing in within the health transformation programme, the back to work plan, and there will be a tailored approach, it does not exist. They do not have time to tailor. They recently got some easements made. They went to the Secretary of State, the PCS Union, because interviews with work coaches are less often and shorter because they cannot cope, their workload is too high. The DWP promised 30,000 staff more staff or something and have employed 5,000 more. They are all stretched to unbelievable levels.

That whole trauma-informed approach should go for them as well, because the welfare of staff is paramount as well as the welfare of claimants. If the staff welfare is not looked at, they are not in a position to look after the claimants and to support them, because you need to feel strong yourself to support somebody who is struggling.

Q102       David Linden: I will come back to resources and transparency in my final questions. Mr Thunder, did you want to add something there?

Jamie Thunder: Yes. Clearly training of work coaches is an important part of it and it is the minimum that a claimant interacting with the Department should be able to expect. However, there will also be a fundamental limit to how much that training can be relied upon to be applied consistently, particularly—as the Secretary of State said to this Committee last week—when there will be 16,000 work coaches as of next year. As well as that extra training, there is a need to see changes made to the business-as-usual processes, to de-risk those processes, not solely relying on training and discretion. That comes down to policy changes that de-risk that as a bright line going through it.

Q103       David Linden: I want to ask a couple of questions about transparency, because for all I might have my political differences with the Secretary of State, it would be fair to say that the new Secretary of State has made much more of a commitment to transparency on the part of the Department. To what extent is DWP adequately transparent about its safeguarding processes and, probably above all, how is it accountable for its safeguarding failures that we have seen and that have been highlighted by Ms Abrahams?

Daphne Hall: There is nowhere near enough transparency. The process to refer into the IPR was one of the questions on your consultation that I googled and I could not find it anywhere. Then the DWP accounts were published. Embedded in the DWP accounts is the referral process for IPRs and how it works—the only place. How are people meant to find that? I consider myself quite skilled at looking for things and it was only by chance that I came across them in the accounts.

People do not know that it exists. There is not a gov.uk page on it even, nothing. We do not see the reports. Trying to get information out of the DWP is like getting blood out of a stone. The CPAG is excellent on FOI requests and then it has to fight the FOI requests and take it to the commissioner because the DWP is putting some regulation in about why it cannot provide that information. My perception is that it does not want to give that information.

Q104       David Linden: If a highly skilled welfare rights expert is struggling to access that, how will somebody who is on the verge of suicide?

Daphne Hall: Exactly. There needs to be a summary of what you can expect from the DWP and what you can ask for. Every claimant should have that right at the start and what they can do if it does not happen. The six-point plan is another example. The only place that is, as far as I know, is every six months the Government deposit the Universal Credit guidance in the House of Commons Library. On rights, they make it slightly more accessible. Nobody can do a search through the House of Commons Library deposits to find that. It is not anywhere else. There is nowhere on gov.uk that talks about the six-point plan; it is only for Universal Credit. I would like to ask the DWP if the six-point plan applies anywhere else, because the only document that refers to it, as far as I am aware, is that document in the House of Commons Library that is completely inaccessible.

There are simple things, and we have said that. We do a lot of stakeholder work. We say to it time and time again why is it not clearly available and why are people not given that information in the format that works for them, whether it is on paper or digitally, just a brief summary for people of what they can do, “We’re here to help”. It is not just, “You’ve got to do stuff for us” but, “We’re here for you”. It is a whole cultural change and it is not going to happen overnight. I completely accept that the DWP has moved and there has definitely been progress with the serious case panel. The fact that it is doing the IPRs is great and that is a movement in the right direction, but it is just a little baby step, as far as I can see. There is an awful lot more to do and a lot of cultural change will be needed.

Brian Dow: To add to that, there are three things here. I agree that IPRs are a very good example of where we are, because there has been a shift and that has to be acknowledged. First, there should be a simple process by which people are informed that if there has been an IPR instituted, that it is happening. Often families and next of kin simply do not know that it is going on. That is something. Secondly, for the process itself it should be straightforward to report it or to be able to ask the DWP to institute one. Thirdly is the slightly more complex business of the publication of IPRs.

There is a really good example. I know that we have talked about anonymity and protecting people’s data and various things that we cannot do, but the Independent Body for Police Conduct’s “Learning the Lessons” magazine is a good example of how we could do this better. It takes cases of misconduct between the police and an individual and explains what has happened, while anonymising all of the personal detail that you can’t put into the public domain. It lays out clearly the lessons learnt and the recommendations. I know that various different police forces have come under a great deal of criticism of late, but that is a good example of how you can deal with something really sensitive where the state is doing harm and publishing the lessons so that as a whole the system can improve.

Q105       David Linden: Excellent. A final question to Ms Hall. There are those within government who believe that the civil service has got too big and that the civil service headcount has to be reduced. Is that compatible with an effort to try to bring in more safeguarding for vulnerable claimants?

Daphne Hall: I think that if you looked after people properly it might create less work. If you support people from the start and enable them to make their claims, they might need less support in the long term if you get things right in the beginning. Benefit decisions is another one. How many go to appeal and then are successful at appeal? If those decisions are made right first time, it can reduce the workload. The attitude is always that we have to cut things back, cut benefits back and cut the support back, but being proactive and getting in there at the start and getting things right at the beginning can free up stuff. It is the whole cultural attitude about how you do it. It probably is a bit incompatible but getting things right early will make a difference.

Chair: Thank you very much. We are getting a little bit behind.

Q106       Selaine Saxby: Good morning. This is the same question that I asked the previous panel on accessibility of benefits. In light of the Chair’s comment, I will focus on the main challenges that vulnerable claimants face when trying to make a new benefit claim. Who would like to go first?

Daphne Hall: I have a shortlist that I will do quickly. Gatekeeping. You are sometimes given the wrong advice on the phone that you can’t claim, so you do not even get as far as making a decision. Lack of accessible information, which we have talked about already. Lack of funded advice. We talked about help to claim, but help to claim now is only online or by telephone. Probably a lot of the most vulnerable need face to face and local connection. With help to claim being as it is, they do not have that local knowledge any more. It goes anywhere in the country. The huge, long wait to get benefits processed. The big one is lack of choice about how to make a claim, whether it is digital or phone, how you have your appointments, whether it is in person, via phone, video or whatever. That is my shortlist to start.

Brian Dow: You have answered with the majority of the ones that I was going to say. I will add that there is an issue with the digital divide. We can think particularly of people with mental illness, but not exclusively, who do not have access or perhaps do not have the cognitive skills to access digital resources. However, I completely agree that the choice of how you are communicated with is important. Then we know that the process of requesting medical advice from professionals is torturous for people. I will also emphasise the point that is true across all conditions, that when people are at their most vulnerable, when they perhaps lack the confidence, the capability and the capacity, asking them to jump through lots of hoops like that, rather than the default of making it as easy as possible, recognising the balance that has to be struck, makes it unnecessarily hard for people.

Jamie Thunder: I agree fully with Daphne and Brian there. I will add that you have mentioned the accessibility of a new claim and it is also worth thinking about the ease in which a claim can end, particularly when it comes to the disability element of Universal Credit and personal independence payment. In effect, you are making a new application every new assessment to keep it in payment and there are lots of ways that can very quickly go wrong and the safeguards there are quite limited.

Daphne Hall: One very quick one is Universal Credit claims that are closed without getting to the first payment. They are closed, a decision letter is not given, their appeal and MR rights are not given and they can read their journal but they cannot write in it so they have lost their ability to communicate with the DWP, other than waiting on a helpline for ages or trying to get into a jobcentre. Claims should not be allowed to be closed. It is not even legally correct just to close a claim but it happens all the time. I saw one on Monday. They said, “We’re going to ring you on the 4th” and they closed the claim on the 2nd without even making the call.

Q107       Selaine Saxby: Are there any other specifics that we have not already covered that you think that the DWP could do to improve support for vulnerable claimants when they make a claim?

Brian Dow: I emphasise the point on how people are communicated with. It is so paper-based and it can take four weeks for people to be replied to. These are people who are not only in medically vulnerable positions, often, but also financially vulnerable positions. When you think about all the things, particularly rocket-boosted by the arrival of Covid, that we are able to do digitally now, it seems to me very antiquated that a lot of this remains fairly paper-based.

Daphne Hall: I will not repeat anything, but reasonable adjustments. The DWP is quite poor on reasonable adjustments. It should almost be a question on the claim form, “Are there any particular things that you need?” with examples given, like, “Do you need face to face, do you need it by phone, can you make it into the jobcentre?” There is a plea that I make regularly, which is to put implicit content back into Universal Credit. That is when an adviser rings and they have not had a formal authorisation but by talking to them it is evident that they have the consent, and that is not available in Universal Credit. Even the explicit consent that they have only lasts one and a bit assessment periods. You have an adviser doing it, then you have to keep putting a new consent form into the journal and it gets lost. It has put in an instant barrier, if somebody does have an adviser, for getting support to help them to resolve the claim. That is a massive problem that we see over and over and over again.

Q108       Nigel Mills: Following on from that, how do you think the appointee system works? Does it work well or could there be ways of improving it?

Daphne Hall: It is inconsistent. I have spoken to somebody who does a lot of corporate appointing. I am talking here about when you have a local authority being appointed. They lack being able to upload documents on to the journals. Sometimes they will make the facility available, sometimes they will not. The post is very unreliable. It would be much more helpful if absolutely always corporate appointees could upload stuff or use an email facility. Sometimes corporate appointees are asked to go into the jobcentre for the ID check, which is just ridiculous. A corporate appointee should not have to attend the jobcentre. There are inconsistencies. Sometimes it is working okay but sometimes not.

Brian Dow: The onus being on the claimant to request an appointee is something that is worthy of consideration. I can see why that is the case but I can see options where the default might be a bit different. To build on a couple of those points, it can take up to 12 weeks to find an appointee. That is a very long time if you are in financially straitened circumstances. There is a particular point where if you are the appointee of a child on disability living allowance, your appointee status automatically ends if the child applies for a personal independence payment when they are 16, unless you apply for it to continue. There is a question about where the default sits and should it be an opt-out instead of an opt-in that is worthy of consideration.

The last point is that it is good that people generally only have one appointee, but if we could have a system where people have two or they are able to work more closely together—which can be very challenging for a variety of reasons—that is another area that could be improved.

Q109       Nigel Mills: There has been some suggestion that it is all or nothing, that if I have an appointee they get to do everything for my benefit claim and I should do nothing, or I do not have one at all and I have to do everything. Is there a way of giving limited appointeeship or limited access so that they could deal with some of the paperwork but not have complete control? I think that that would be of benefit. Something that is slightly strange is if I am applying for multiple benefits, say I have an accident and have to apply for UC and PIP at the same time, I have to do two separate appointeeship applications, even though I am one individual who is clearly going to need help with both. Would a system where I have it for all benefits that I might need to apply for be progress? You all seem to be agreeing.

Daphne Hall: Yes. It only happens occasionally, but somebody who did have an appointeeship, I think it was for PIP, then the claimant without capacity put in a UC claim when they should not have done and killed the legacy benefits. There almost should be a safeguard there that if somebody has an appointeeship and they put in a claim themselves—I do not know why they did it, somebody told them it would be a good idea or something—if there is an appointee maybe that should be checked before the claims process because of the impact of doing that.

Q110       Debbie Abrahams: My questions are about preventing harm. You have said a lot already, but based on what you have said and where we are now, what else do you want to add to where we should be in a safeguarding system? You have mentioned that a policy perspective around impact assessments might be one aspect, around a duty, but we need to define who that would apply to and the different levels in the system. Those are important points that Ben made. What do you want to add to what you have already said?

Daphne Hall: Starting at the frontline, it is even what the first contact with a claimant looks like, starting that relationship. The DWP is not seen very well, generally, by claimants, and that first contact can make such a difference. If it was focused differently, that would make a difference. How you deal with a person is so fundamental to them feeling open to disclosing things, because if you do not trust somebody you are not going to disclose.

Brian Dow: I will add a few things. The first is the quality of collaboration between the DWP as a whole and other Departments. We have seen some very terrible cases and you will be familiar with them.

Debbie Abrahams: Is that at a local level?

Brian Dow: I am thinking particularly at a local level, where you have an individual who might be under the care of the local authority and receiving housing benefit. You will be familiar with a couple of very tragic cases recently where the communication between the two—the DWP perhaps not informing the local authority of the situation—has led to a series of events that ultimately has resulted in some cases in suicide. That can definitely be improved.

There is how information is stored. This is true in other areas of public life, that the desire to protect confidentiality can make it difficult and you can end up with an unfortunate outcome where people are jumping through hoops every single time. Store information in a way that is confidential but prevents claimants having to tell the story again and again and potentially being retraumatised. Of course there is a wider issue on where sanctions are applied and for whom.

Debbie Abrahams: I was going to come on to that, on the policy side.

Brian Dow: I thought you might.

Q111       Debbie Abrahams: Jamie, do you want to add to that?

Jamie Thunder: A commonality between the awful cases that have in large part prompted this inquiry is the many ways in which someone’s benefits—and they are subsistence benefits—can be stopped or reduced substantially or entirely and very suddenly. Whether or not that decision is correct—and very often it is not and it could be overturned later, which is of no use to you in the moment, but even if that is correct it can have a big impact in exacerbating existing vulnerability or indeed make you newly vulnerable. For example, if you are reassessed for the work capability assessment and you no longer get it and that decision is correct in law and has been correctly applied, you might have been relying on that extra money to top up your rent but it goes instantly. You have no real way of adapting; you are not going to find a job instantly. That can make somebody newly vulnerable. Those cliff edges—and there are many of them in the system—pose a real problem and it is something that would be very helpful for the Committee to look at.

Daphne Hall: There are the additional support markers that they have on PIPs that flag that somebody is vulnerable and it means that their claim will not be closed if they fail to return a form. It is sort of a watermark so that you see it instantly. I would like to see that across all benefits, particularly Universal Credit.

Jamie Thunder: That only applies if you do not return the form. If you do not return the assessment, it will still stop.

Q112       Debbie Abrahams: We have a safeguarding policy and system that the Government would like to introduce to the Department for Work and Pensions but we also have, within that Department, other policies that may not lend themselves to a safeguarding approach. There is a policy dissonance between trying to make sure that we enable, as Daphne described, and what do you want in how we can help you to get into work. How can we help to get you better and what do you need from the health service to get you better, if that is a potential? How can we do that? Then we have a sanctions policy, we have issues with some of the practical sides of policy, the assessment process and so on, where, as Daphne mentioned, you get the knockback but on appeal you get there. How can we square this circle?

Brian Dow: How long do we have? What you are pointing to there is that it feels sometimes—certainly in the work that we have done with people who have severe mental illness we hear this repeatedly—that bits of the system feel almost as if they work against each other. You mention the question of appeals and how many people succeed on appeals. That would suggest that the first bit of it is not working correctly. That then begs the question about the motivation.

Sanctions are another good example. There is no question that some sort of sanctions regime within the benefit system needs to exist because there will inevitably be people who will take advantage of it, so you need to have the right incentives to move people. However, is that squared against the reality that for the majority of people—again thinking particularly about mental illness here—who want to work in some way? It does not feel as if the benefit system as a whole is set up like that, so I think that you are right.

That comes back to the original question. The notion of a statutory duty of care to me is the moral floor, or the legal floor, but the moral ceiling is having a culture of constant improvement where you are learning from the mistakes that are made, where you are open about them, where there is a level of transparency and that the harm, and in some cases the death, that is caused by the benefit system—taking the point that the National Audit Office made before about the sheer numbers of people, but one death is one too many—results in a much more coherent approach to the balance that has to be struck.

Daphne Hall: Picking up on the sanctions, something is going so wrong there where 97% of sanctions are for failing to attend an appointment. They are not only for not trying to get a job but just that they missed an appointment. Something is going wrong there, because it is not saying that they are not trying. Maybe they have had a terrible day, maybe they missed the bus or whatever it is. Sometimes they are given for literally being an hour late. There needs to be research on what is going on there.

I do not know whether you saw the stats that came out yesterday about the numbers, 35,000, who were only on the standard allowance were still sanctioned for six months, because failures to attend an interview are open ended. Why are people not re-engaging? Something is wrong. I do not believe that there are 35,000 claimants out there going, “I just don’t want a job, I don’t want to work”. It is not that, it is vulnerability, I am absolutely certain, in 99.9% of those cases, so research needs to be done. The sanctions policy is not working. It is just penalising vulnerable people and leaving them destitute, beyond destitution, and that is a problem.

Jamie Thunder: That is an important point. The sanction has an immediate and very significant impact. If you are a single person, it would remove your entire living money, your standard allowance within Universal Credit. Reducing to that level means that you cannot possibly meet your energy needs, your food costs, your rent, if you are needing to top that up. That means that you are vulnerable and you are vulnerable because of the very instant and very severe impact that a sanction has. That means that there is an inherent and basically unavoidable risk at the moment of the policy causing huge harm. Even if you have fantastic systems—and there is the WCA and there are easements available—and those systems worked perfectly when people apply for them, vulnerability will not always be known and any system needs to take that into account. At the moment it simply does not.

Q113       Neil Coyle: Daphne, you touched on stakeholder engagement work earlier. I wonder, since I was involved in it, if the DWP has got any better at engaging with external organisations on this issue specifically about safeguarding.

Daphne Hall: Our stakeholder work has improved and we have a lot more contact and we have a regular inbox to go to. Sometimes it works really well. If we identify a systemic issue and things are not working, we have an email inbox and I usually get a reply pretty swiftly, within a week or two. Often they will have taken action and will have investigated. I might have a few cases and they will have investigated them and have issued new guidance or training for staff. Quite small scale, we have some good things like that.

The move to UC meetings—and we have a lot of them—are very good in lots of ways and we have made some progress, but time and time again we are saying that something is going wrong here and 28% of the tax credit claimants have not migrated safely. We need to know why before we scale up. They are still ploughing ahead and saying that they will have another 500,000-odd done by the end of March. That is about another 100,000 claimants a month. If 28% of those do not go over, what is happening to all those people?

The average loss of income to those tax credit claimants is £300 a month, so we are not talking tax credit claimants who were only getting £20 or £50 a month and they can manage without and do not want the hassle, the average is £300. It is huge amounts of money that those claimants have lost. That is where I find that we are hitting the brick wall on those big issues. On the small scale there is lots of good work. As we were saying before, with 10% not getting timely benefits, a quarter not migrating safely is horrifying.

Brian Dow: I think that the quality of stakeholder engagement has improved significantly. I had a meeting yesterday with two very senior officials at the DWP. I had best not name them in front of the Committee. It was clear that they do want to work with the wider sector much better. I pay tribute to Tom Pursglove, because Rethink has been running a campaign, as some people will know, on the effects of benefits deaths. We have tried very hard to get previous Secretaries of State or Ministers to meet with the affected families and Tom Pursglove did meet with them recently. Therefore, I do sense a desire and a recognition at both a political and a policy level that if the system works better with stakeholders you will get better results.

Q114       Neil Coyle: That is positive to hear. Is the positive movement based on individual cases and some of the publicity or is there now a search for solutions or the means of trying to avoid for specific groups of vulnerabilities, potentially?

Brian Dow: That is a great question and in a way quite a hard one to answer on behalf of the system as a whole. My personal take on it is that, a bit like you have had in the health system—and obviously I am looking at it from the mental health perspective—there is a very strong level of collaboration between various different sector partners and the system as a whole. Arguably, they are part of the system. We have a common and shared aim. I have talked here on a number of occasions about the very difficult balance that has to be struck between preserving the rights of individuals and the duty to the taxpayer.

I sense that what is happening is a recognition, particularly when you have significant problems of economic inactivity—a phrase that I must say I hate—that we have a common goal in supporting people back into the different types of work that are available in different flexible fashions. I hope that it signals something that is much more long term.

Q115       Neil Coyle: If anyone else wants to come in on that, that would be great, but are there specific recommendations that stakeholders or other agencies or organisations working with DWP have flagged up or tried to get taken on that have not been so far, that have not been shared in advance by organisations or could be?

Daphne Hall: Sorry, could you say that again?

Neil Coyle: Are there recommendations that you have made or Rethink has made or anyone that the Department has not taken on board yet around specific groups?

Jamie Thunder: I think around managed migration, as Daphne mentioned. There has been a number of recommendations made about not scaling up as quickly as the Department—

Q116       Neil Coyle: For specific groups of people who may be considered vulnerable under the way that DWP logs these things, or just general policy?

Daphne Hall: We have not got that far along.

Jamie Thunder: General policy.

Daphne Hall: On the move to UC, we have had some good collaborative meetings before it brought couples on, listening to our concerns. I have to say that it has been very good and listened to our concerns. We have had some very good stakeholder collaboration. There was one thing that it was going to do—where we had a confidential meeting, so I will not say what it was—and we all said, “Please do not do that”, and it took it and it did not do it. Some of the officials are absolutely excellent and they actively seek our opinion, which is good.

Yes, there are things where we do not feel that we are making progress. Transitional protection is another one and people not necessarily getting the right transitional protection. However, it is a mix. Sometimes we are struggling to get there but there is a lot of good movement as well.

Q117       Neil Coyle: When the Department makes payments to landlords and others in the trusted partner status, would some trusted partner status be useful for organisations raising specific concerns or making specific recommendations about particular groups of people? Would this be a means, even within a broader policy, of identifying specific vulnerable groups that could benefit from a different approach that could be tested with a reasonable adjustment perhaps?

Daphne Hall: I always feel that with the stakeholders we have that option. Sometimes there will be things that we bring up for specific groups. Is that what you are getting at? That is open there. We can bring up whatever we want. If there are particular groups that we feel are—

Q118       Neil Coyle: There is raising it and then there is getting the concern taken on board as quickly as possible to protect people from losing out or coming to harm. It does not feel like that is there at the moment. It is how to build something that allows for better protections going forward.

Brian Dow: A good test for us is that we have just been, as I said, granted some money by the Lloyds Bank Foundation to mobilise the voice of lived experience to help improve the WCA process. I think that our aim ought to be not just to protect the most vulnerable but to make sure that as a whole—coming back to the duty to the taxpayer—you have a system that gets to the heart of who is able to do what. At present it is not. The noise from the DWP is that it is very positive about working with us to use that lived experience. I am realistic about that. It does not mean that every single person will get what they want, but if you end up shaping a system that interacts with people much more effectively, it is better for everyone. Therefore, I am hopeful about how we might work in the future.

Q119       Neil Coyle: The Department announced the extension of conditionality and sanctions. Earlier, Daphne, you said that a statutory duty should look at safeguarding terms of impact assessment policies on how they safeguard or how they might protect some people at risk. The equality impact assessment system is meant to prevent public policy change that would harm disabled people. Disabled people for the purposes of this would meet vulnerability tests. Is it that the Department is not using the equality impact assessment properly or is it that the equality impact assessment needs improving going forward to better strengthen how organisations like yourselves can say, “This policy would have this detrimental effect and this is how to mitigate it or not do it”?

Daphne Hall: For example, the EHRC bought up that there had not been equality impact assessments done on the recent changes to the work capability assessment. It has not gone ahead yet but it has said that it is going to go ahead and we still have no equality impact assessments. That is a major concern because there are huge issues, particularly with the substantial risk changes and stuff like that, where vulnerable people could find themselves falling foul of the changes but we have no equality impact assessment. It is a problem.

Jamie Thunder: If we do not have an impact assessment when we are responding to the consultation, that greatly limits our understanding of its potential impact and the way in which we can then engage with it. It was very disappointing that that was not published earlier.

Q120       Neil Coyle: We are short of time, but without the equality impact assessment or some safeguarding mechanism, is there a risk that those plans would exacerbate pre-existing vulnerabilities or retraumatise certain people within the benefit system?

Jamie Thunder: Yes, and with or without the impact assessment that is a risk of those policies.

Siobhan Baillie: I think that you have answered all of my questions so I am going to give you a break. I was going to look at internal process reviews but we have talked about that a lot.

Q121       Chair: Can I pick up the point that you were just making about the recent change to the work capability assessment, Daphne? We have also had the announcement of the chance to work guarantee recently. From what you said about the WCA, you do have some concerns about the vulnerable. Tell us briefly about those.

Daphne Hall: Major concerns. The estimate on the policy documents with the autumn statement is that they think the caseload on LCWRA, the limited capability for work and work-related activity, is 370,000 fewer once it all comes in. That means not only a halving of their personal living money but also the conditionality. We have talked already about the risk of conditionality. These are some of the most vulnerable people.

I am concerned about all the changes but the substantial risk one particularly. That is the main way for getting into that LCWRA group, particularly for people with mental health problems, because they do not always neatly fit into those points. That was the safeguarding thing. We do not know yet what the criteria are going to be, but clearly the intention is to tighten it massively so that a lot less people get in. What is going to happen to all those people? I have worked in advice work for a long time. Some of the people in that support group are incredibly vulnerable and there are huge concerns, and that assessment has not been done.

On the chance to work, that is the one positive thing in there, because they have said that people already with their LCWRA will not get reassessed. I wish that they would do that to everybody. That would enable people to look for work, feel safe and not worry about it. Once you have your LCWRA status—unless it is one of those very short-term things like pregnancy or cancer or whatever—just leave it there and say, “Yes, you can try work”, so that is good.

It is difficult to comment on properly because we do not know the details of what they are going to do, but the fact that they are going to try to reduce the caseload by that amount is hugely worrying because those are not people who are in a position to be storming out to work or whatever.

Brian Dow: I completely agree with that, obviously. That distinction is hard to reason, frankly. Within the chance to work guarantee there are some very positive developments, particularly the potential expansion of IPS, which is a good, well-evidenced programme. We have been working with a commercial partner to expand the IPS model so that it supports the employer as well as the potential employee and looks at the broader range of factors that can cause them to drop out of work. Hopefully that might be part of it in the future. Therefore, there are things within the overall programme that we should be pleased to see and that do illustrate, to my mind, a growing understanding of the separation between those different groups.

Jamie Thunder: I completely agree with Daphne and Brian on the substantial risk. Simply redefining the risk does not remove the risk in practice and it is very disingenuous to claim that it does.

I want to mention some of the longer-term reforms that the Department is looking at that were announced in the White Paper last year on the eventual abolition of the work capability assessment. One risk that exacerbates is that it would, for nearly all clients, tie the receipt of extra Universal Credit to the receipt of PIP. That would mean that a disallowance of PIP could now in total, across those two elements, mean over £1,000 a month drop in income overnight. It is very hard to see how that would not or could not make somebody vulnerable.

Currently, although nobody likes the treadmill system of multiple assessments, it does provide a bit of insurance against one of them going poorly. The abolition of the WCA will also make work coaches responsible for determining how somebody’s health affects their ability to undertake work search or work-related activity. That again carries a big risk of poor decisions. If it is in the claimant commitment, there is no formal challenge route. You do not have the same MR and appeal because it is nominally a jointly produced document. That, of course, could also worsen people’s health and further expose them to conditionality and sanctions. Those are things that we are very concerned about in this area as well.

Chair: Thank you all very much. That concludes our questions to you. If there is anything that we have not raised with you or anything that occurs to you afterwards that you felt you ought to have said to us, please do email. We would be very interested to hear any further reflections that you have. We are very grateful to you all for joining us this morning. Thank you for coming and answering our questions. That concludes our meeting.