Welcome to this oral session of the Work and Pensions Select Committee. We are delighted to have with us today the chair and members of the Social Security Advisory Committee, Stephen Brien, Carl Emmerson and Bruce Calderwood. It is very nice to see you here. Stephen, as chair of the committee, do you want to make any opening statements?

I will just say that we are delighted to be here and welcome the opportunity to share our ways of working with the Committee. I know that you have several new members and we look forward to the opportunity to exchange views.

Thank you. Would you like to start off by telling us about the role of the Social Security Advisory Committee, how it is set up and what its responsibilities are?

The Social Security Advisory Committee was set up 45 years ago this week or this month, in 1980, so we are celebrating an anniversary at the moment. It had two main purposes as an independent committee advising the Secretary of State. One of its roles is to scrutinise the majority of secondary legislation with respect to social security, and the second is to engage in advice and assistance to the Secretary of State, which is more self-generated. We act reactively whenever regulations come to us for scrutiny. We meet monthly, 10 times a year, and have quite a number of ad hoc meetings to receive officials and scrutinise regulations. Separately, we take on the role laid out legislation to advise and assist the Secretary of State, which involves us identifying areas where we think further research is needed or policy has been neglected, and we provide independent advice to the Secretary of State on those issues. Our advice can range from fairly serious, long, formal reports to shorter letters or even meetings I may have with the Secretary of State or Ministers on specific issues.

You been in post for five years since your first appointment. What key lessons have you learned during that period?

The first lesson is about maintaining good relationships with officials and Ministers. For advice to be useful, it needs to be heard and to be able to be heard. I have found building trust, not only with the Secretary of State, who is the person we are charged with advising, but with his or her officials, to be very valuable. As you will be aware, a lot of the role is about discussions, negotiations and making sure that we get access to information, and that when we provide advice, we are doing it in a way that the Secretary of State can listen to. You may remember my appearance in front of this Committee before I was appointed. I said that I was looking to engage in such a way that sometimes the Secretary of State will say, “Thank you, that is very helpful,” sometimes, “Ouch, we didn’t know about that, glad to hear about it,” and at other times, “Thanks, but actually I have decided to do something different.” I think it is important to be able to have a dialogue around all three layers of that spectrum.

You have been reappointed for three years, so congratulations for that.

Thank you. Correct, yes

Is there any reason for that?

I am not aware of a specific reason, I can surmise, but I am very happy to be reappointed none the less.

You have covered the working relationships and their importance in formal exchanges and behind the scenes. How frequently do you meet with the Secretary of State?

I meet with the Secretary of State probably once or twice a year, but I also meet with relevant Ministers. Typically the Minister for the law is the sponsoring Minister and I would meet with them every six to eight weeks. At the moment I am meeting with Minister Timms quite regularly, given that he has a very specific social security brief. There is a very strong overlap between our two scopes of activities.

Fantastic. Within your responsibilities and terms of reference, there is an opportunity for you to provide unsolicited advice. What are the circumstances around that part of your role?

Sometimes there is a long, drawn-out, rigorous process where we identify an area. I will take the example of an area that Carl Emmerson led on, looking at new-style ESA or JSA. Following on from the development of universal credit; the contributory benefit system needed to be changed and adjusted so it would dock with UC. That happened 10 years ago and there had been very little policy development in that space since. As universal credit evolved, as the benefit definitions evolved, and indeed the support for UC recipients evolved, we felt that the contributory benefits had not gone along the same journey, and because there had been low numbers this was a neglected area of policy. Post-covid—it became very clear during covid that a very different tranche of people needed support from the state, and the contributory benefit caseload expanded—it became much more important to have a really good system, to think through the learnings from universal credit and to apply them to the contributory benefits. Our work on that was one of the more serious, significant pieces of work that we reported on to the Secretary of State. I am delighted to see echoes of our recommendations coming through now in recent government proposals.

Indeed. Is there anything that you are thinking along those lines at the moment?

We are working on a number of areas. We are looking at how universal credit supports younger people, particularly as people make the journey from age 15 or 16 from being part of a household on universal credit, part of the benefit unit, through a journey of going into either further education, college, apprenticeship, sixth form, on to third level education and so on. Their role in the household shifts, the rules for benefit shift and for other kinds of support also shift. We want to make sure that that journey is as smooth as possible, but also that the system is not simply a level playing field but helps make sure that young people’s choices are as balanced as possible, so that they can make the right decisions for themselves and for their long-term future without being bound by benefit rules or immediate financial impact on the family. For example, if somebody goes into further education, they can continue to get support, child benefit and so on. If they take an apprenticeship, they lose some of those benefits. There are incentives in the system against apprenticeships. It is not about the committee’s view on whether apprenticeships are good or bad; we always take the Government’s policy intent as our starting point, and when we see a disconnect in the system between what the Government are saying they want to do and the signals they are sending from the machine, part of our job is to call that out and say, “This is an area that we would recommend the Government act on.” We are just working through that at the moment, so we do not have any specific recommendations yet, but we are taking evidence on it. Some of our colleagues and I had a very good visit to Central England last Friday. We saw the careers service and met with work coaches, careers advisers and so on to help us understand how the system works on the ground. We found that very valuable in giving us more tangible evidence to connect with our deep scrutiny of the regulations.

Thank you. I will hand over now to David Pinto-Duschinsky.

Thank you all for taking the time to come to speak with us today. I will focus on the committee’s scrutiny work. In 2023, there was a review in which DWP recommended that the committee clarify its remit for the operational impact of legislative change. What has been done since the review to bring that clarity?

There have been some conversations internally. We had that recommendation because there was a discussion around the extent to which it was appropriate for us to be getting involved in operational issues. Our remit is predominantly the legislation. As you can well imagine, however, to know whether a piece of regulation is serving the policy intent, we need to think through the value chain of how the legislation is implemented on the ground. As a committee, we have remained quite firm in our view that we need to be able to understand operational implications. We need to be able to engage with work coaches and others and those in service centres elsewhere to understand how the regulations are implemented and what the experience for claimants is like. In the last period we have put together a very good set of visits. After a period when we were not getting great access to staff members, I am certainly seeing a change within the Department now in being able to set up outside site meetings. Friday’s trip is evidence that dialogue on operational issues is now easier. We regard it as very important, not because we want to stray outside our remit but because we need to understand the downstream consequences of the areas we are scrutinising.

When we think about legislative impact, we think about primary and secondary legislation. The review also suggested it could be useful for the committee to be sighted on certain aspects of secondary legislation within six months of the primary legislation, to ensure that the impact has been fully considered. Are there any examples of where that has taken place? If so, has it proved useful?

I cannot give you any examples where it has taken place since then because there have not been any areas that we might have probed in that way. We are in discussion with the Department and we are fairly confident that we have sufficiently good relationships that they will discuss areas of secondary legislation with us that they are not required to in an informal way, so that we can pick up weaknesses or issues that they may not fully have bottomed out before it gets to the point of coming to Parliament in the regular way. However, there has been no significant opportunity since that particular recommendation.

We could have something over the summer. We did have extensive briefings from officials on the Government’s proposed welfare reforms and we definitely touched on areas that they were expecting to come to us. This is why I made the point earlier about the emphasis we place on managing good relations with the Department so that we are able to be the internal challenge team—so that they can speak to us in confidence and we can give them genuine advice over and above what we are required to do publicly through our formal scrutiny.

Another aspect is the urgency procedure. Under what circumstances would you consider it appropriate for DWP to invoke that procedure?

The starting point has to be genuine urgency, predominantly a function of external events rather than a Government’s internal dynamics, because internal dynamics are self-created. Looking back over the last five years, there are three obvious and clear reasons why urgency was needed and appropriately so. Covid-19 is the pre-eminent example. We had a huge number of emergency regulations and the committee worked as fast as it could either to quickly review and scrutinise or to accept with good grace the fact that they had to be done through urgency. The second example is the downstream consequences of EU treaty regulations where, again as a result of certain agreements, things have had to be done at very short notice. Again, these are externally driven reasons. The third major category has been international events where there have been reasons to change the eligibility rules for the habitual residence tests—for example as a result of events in Afghanistan, Sudan, Ukraine and so on. Those are three very clear and obvious categories where for outside reasons the Government must act expeditiously. There are other times where the Government feel they want to get something done quickly. You will know from various letters that I have written to a number of Secretaries of State that I have pushed back and said that we did not feel that the urgency provision itself was needed. If the Government are in a rush, we have always made ourselves available at short notice to have meetings beyond our monthly cadence of meetings. We have always made ourselves available to have short-notice meetings for the process that has been set out in law to be followed as expeditiously as possible without needing to invoke urgency and circumvent the normal procedures.

Beyond writing letters, do you have any other recourse if you believe the urgency procedure has been invoked without adequate cause?

We have no direct recourse on that. If urgency is invoked, the regulations will get laid without our scrutiny in advance. However, we do scrutinise these regulations after the fact and we still would have some powers to engage in formal reference. That would not necessarily be anything more than the power of the pen, but we can highlight issues and insist on the Secretary of State responding to us. Beyond that, we do not have powers. Parliament is ultimately sovereign, so we cannot stop the process.

A final question. Does the way that regulations are presented to the committee give you adequate scope and time for effective and genuine scrutiny?

Generally, yes. Carl, do you want to pick up on that? We will be meeting again today because some regulations last week were not as well prepared as needed, and we will take multiple bites of the cherry at it if we need to. When urgency is not invoked, the power of the pen that we have is to take as much time, within reason, as we feel is needed to get them right.

When it is working well, which it often is, we do. Sometimes if a particularly complicated set of regulations is coming up, the Department has briefed us in advance of coming to us for the formal session. That allows us to highlight areas that we want to ask questions on. It also allows us to make sure that we are on top of the issues. Just hearing them described more than once in advance of getting the papers can help quite a lot. Then a week before the main SSAC meeting a subgroup will meet and discuss the main issues, make sure that we have a view on what the key things are that we need to interrogate with that set of regulations, and feed that back to the Department so that it knows the areas that we are likely to be concerned about. That is a subgroup of the committee. I am not on that subgroup, but I have always found it useful to go along and observe the interaction to help get my brain around exactly what the regulations are supposed to be achieving and the concerns I might have. We get all the paperwork a few days in advance so that there is time to digest it. The other thing that has helped recently is that if you go back to 2022, 2023, the committee was very small relative to how many members we should have, because of decisions made by Mel Stride when he was Secretary of State. Under the new Government we have been able to recruit new members and we are now back up to a better size of committee so that we have more capacity around the room in those meetings.

Where it can become difficult is if there is a fixed deadline and the Government have a deadline that they cannot do anything about; they have to deliver. We will assist them to avoid invoking urgency by potentially arranging a meeting, but then we have a decision to take as to whether we have enough information to respond. Sometimes we do. For example, earlier this year we were looking at regulations that changed deductions for child support, and there was a very limited time available. The Government had to regulate to ensure that lone parents could continue to receive child support payments. We thought that we had enough time to come to a firm decision and we took what we call formal reference, which means we write a formal report. Normally, we would go out to consultation. We were sufficiently confident in our judgment that we could do it and we completed the process in that timescale. Sometimes it does not work out that way because the regulations are sufficiently complicated. There may be issues that we do not feel we can bottom in the time. For example, with the winter fuel payment regulations we might very well in normal circumstances have decided to take it on a formal referral, and gone out to consultation. That would mean either the Department would then need to invoke urgency, in which case our report would land some time after the regulations were laid, or we would write setting out our concerns, but we would not necessarily have bottomed all of them. The choice that we made was to write setting out our concerns. Had we gone out to consultation, some of those concerns might have been mitigated, some of them would have gone away and others would have been broadly substantiated, but we did not know at the time that we wrote that advice.

To go back to your earlier question about the six-month rule, this is not a six-month rule issue but it is analogous to what is brought to us and what is not brought to us. The deductions regulations that were brought to us were about a change in the deductions order for universal credit, and that was needed because of a separate regulation that was not brought to us. There was a two-step process. First it was regulations that limited the total amount of deductions, and then a bit of an “ah” moment that if that goes through, some of these deductions, which are not simply to utility companies, court fines, but are transfer payments between one household and another through the child maintenance system, could get squeezed. The consequential set of regulations then became that these are going down the track. They are already in place. We now have to remediate them over here, and that is what created the deadline. Ideally, both sets of regulations would have been developed at the same time and both sets would have come to us. We would have treated them in the round, fully recognising that, in theory, half of them were not obliged to come to us because they were simply a change in the amount of deduction, but when you bring the two together they are a substantive piece. There are other situations like this as well. It is often when there are these combinations of regulations that one bit has been fixed and the other bit then has to get adjusted. That sometimes creates urgency because then there is a deadline because something has already happened that they then have to correct for.

Can I follow up? Bruce, you chair the scrutiny committee or subcommittee. I want to understand that process. From what you are saying, you will sometimes be given relatively generous timescales to do this, and in other cases, you will have to quite quickly scrutinise the regulations to decide whether you will do any work on them. Is that correct?

Broadly, yes. What happens is that if all goes well we will meet a week before the main committee meeting to look at the regulations that will be presented to the main committee meeting and have a set of papers that will contain the draft regulations, an explanation of them, an equality impact analysis and various other supporting bits of information. There are two decisions to be made. One is: does this need to come to the main committee meeting or is it technical, raising no new policy matters of interest? It might just be correcting a word in a regulation. It would be a waste of time for officials to come to us and for us to talk about it. We decide whether they need to come at all.

Do you use any screening criteria?

Well, the screening criteria for that are: is it non-contentious; is it technical; are there any losers; does it raise any points of policy interest?

Does it relate to other policy that is also being reviewed, given what has just been said?

Yes. It is open then to the main committee to say, “We think that you have made a mistake and that we ought to have seen this because it raises issues that require further discussion.”

Did the UC PIP Bill come to you?

The Bill, no.

So the Bill would not?

No, and anything of that order we would definitely not regard as a technical change.

What about the Right to Try regulations?

We assume that the Right to Try regulations will come to us; they should come to us. They are a substantial change in the way in which the law works, so it would be very unlikely for us to treat them as something we can clear.

Did you see the Right to Try draft regulations—are they draft regulations?—that accompanied the UC PIP Bill and that were deposited in the Library?

We have seen the latest version of the regulations and we expect them to be brought to us, formally speaking.

Is there more than one version? We have seen one version, we have not seen the second.

This is a good example. We maintain a good day-to-day connection with the officials and this is a classic example of what Bruce said about the pre-briefing, informal discussions as we are warming this up, making sure that we are aware of what is going on. However, we very definitely do that as informal conversations where it is predominantly making sure that we are tracking. We have not started any scrutiny on those yet. We are merely in our own evidence-gathering, informing mode.

That is very helpful, thank you very much, on the day-to-day practicalities of your committee.

Yes, and Carl will undoubtedly come in as one of our more enthusiastic scrutineers of that area, given his professional experience. We have found that it is very variable quality. Some teams of officials take it very seriously, think it through very carefully and come to us with very impressive, very thoughtful impact assessments, and we find them incredibly useful. At other times, there are two lacunae: one is that they come to us with very little data or an explanation that they cannot do the impact assessment because of a lack of data; at other times we get reams of data tables but no thinking behind it. One of the things that I am in an ongoing conversation about with Secretaries of State, the permanent secretary, officials, chief analysts and so on is to make the argument that this is a critical part of the policymaking process and it should not be regarded as the stepchild to the EM, the explanatory memorandum. What you see, talking to officials, is a real sense that they have to get the explanatory memorandum very clear. They feel like they are on the hook to have a good argument to be able to explain that. But they do not feel or they do not demonstrate the same level of accountability for the equality impact assessments. Carl has been doing a lot of work on this side and I will let him expand a bit more. If necessary, we can come back on it.

As Stephen says, to do them well requires first thinking through whether this directly discriminates in any way, and then thinking through why it might cause indirect discrimination in any way, thinking about protected characteristics and, in some cases, thinking about other characteristics that we might care about. That is an intellectual exercise that needs to be gone through that can help policy development. Then there is a question about what data we have that can shed light, or if we are concerned that something might discriminate in some way that is not intended, perhaps there is some data we can get that will show us that there is not a problem here. All too often there is a presumption that there is no discrimination, or, “The administrative data does not tell us what we would need to know, therefore there are no data, and besides it does not discriminate, therefore there is nothing to see here.” If you take the Government’s decision as of a year ago that only households receiving pension credit should be entitled to the winter fuel payment, we were not presented with any details, for example, on what impact that might have on poverty. A decent assessment would think carefully about pensioners who reached the state pension age before and after 2016, because that determines whether you are entitled to the new state pension. You would want to see analysis split by that. The Government’s intent clearly was to make sure that low-income pensioners got the winter fuel payment, so you want to think through carefully who is entitled to pension credit but not claiming it and whether that might be different by ethnicity or by disability. But then what about other criteria? What about how digitally enabled the household is? That is not a protected characteristic, but you might worry about it. What about different regions of the UK? Again, region is not a protected characteristic but you might worry about that too. We did not see any of that thinking at the time. This year, when the latest change to winter fuel payments came along, the assessment again seemed to be rather lazy in its thinking. It asserted that automatically all pensioners would get the winter fuel payment. That is true of pensioners who are receiving the state pension, I believe, but some may have deferred the state pension. They will have to apply for the winter fuel payment, so as a statement of fact that was incorrect. Then going through the protected characteristics, the first thing you can say is that rather obviously the policy discriminates by age, because if you are not a pensioner you are not going to benefit as much from this policy as if you are. So you need to be clear about that and say that the Government are okay with that form of discrimination by age and that is the purpose of the policy. The evidence presented to us on disability was very weak, not using the official definition. They tried to use some proxy data but did not set out clearly how that was working. On the other characteristics, again there was a presumption, “We don’t have any data, therefore there is not an issue.” One of the protected characteristics is pregnancy. I do not think I need very much data to say that the winter fuel payment expansion was not going to do much to help pregnant women, and a quick Google search will reveal that there has been one woman in the UK who is given birth over the age of the state pension age. So you do not really need administrative data, you just need to be clear that, for better or worse, the winter fuel payment expansion does not do anything to support pregnant people. They had not gone through that thinking, which is quite disappointing because it does not take a lot of resource. In that case, I do not think that it is anything that the Government would worry about. We would not need you to change policy in any way, but in other examples you might think that maybe there is a policy mitigation that you want to think about, the way in which the policy is communicated, the way in which it is described to people or perhaps to those administering the system that could help alleviate some unwanted discrimination that might otherwise occur.

There is a number of reasons. In a number of situations detailed impact assessments might not have been helpful to a political agenda. That is just a statement of reality. That should be no reason not to do them. While that certainly may well be the case occasionally, I do not think that that is the driving force behind it. I think that there is a broader issue of capability and mindset at the official level, the mindset that I referred to earlier about the difference between saying, “The explanatory memorandum is really important, I have to get it right and I have to be on top of it before I come to the scrutiny” versus the impact assessments, which seems to be the stepchild of that set-up. I also think that there is an issue around capability, not necessarily in a technical sense, because often when we probe and push and question we find that a lot of these answers are sort of scattered around among the officials. However, there is the discipline and rigour of saying, as Carl has just laid out, “Let’s just take half a day. We don’t need any numbers. Let’s just work through the logic of this,” because we have had to work through most of the logic of this to think through a good set of regulations anyway. If you have prepared a good set of regulations, you have already done 90% of the thinking needed for the impact assessment. The last 10% can be enormously confidence-building that you have the regulations right or prompt, “Watch out, there is something you have forgotten about.” We would love to see, as a committee, the logic of the impact laid out and then, nearly at the end, some quantitative validation of the scale of the various issues, but the issues themselves do not need the analysis. We find that most of the work of the impact assessment is handed off to the analytical teams. My sense—and it is a gut feeling—is that they are doing their best with the briefs that they get. They are analysts and have access to administrative data predominantly, so they will see what administrative data can be used to support the impact assessment. If the onus and the ownership of this issue was much more on the policy level, first they would have to lay out the logic, as Carl has done. Secondly, they would recognise, because they have internalised the issues, that with the Google example, you do not need to go into administrative data to solve 50% of these issues. Common sense, common knowledge and publicly available information will do a lot of it. Then there are some specific issues, for example dealing with apprenticeships, a policy that affects apprentices. We would need to understand the protected characteristic make-up of apprentices. We will not have much of that in the public domain and we will need to delve into admin statistics to get our hands on that. But we will not get a perfect picture in admin statistics either, therefore we need to lay it out and show what is proxy, what is real. That does not take a huge amount of effort and, if done well, possibly reduces the workload. It means you are doing a lot of higher-order thinking and then more discrete analysis rather than lots of detailed analysis that loses its shape.

As Stephen said, if you have done the job properly in the first place, the impact assessment should fall out of that. That was the original intention behind the requirements around equality impact assessments, that before you take a decision you have gone through that thought process so that in a sense you then document it rather than taking a decision and then doing it. Mostly officials and Ministers go through, because obviously they have to do it, “What will be the effect of this? Who will benefit? Who will be hurt? What are the unintended side effects?” It ought to be much easier than it appears to be to extract that thinking process and do a little bit more work to flesh it out with any additional numbers that might be needed.

Carl, you have started to answer this question already.

I think that the key is, as the policy is being developed, to sit and think through protected characteristics and other things that we might be thinking about or concerned about. It is probably quite easy to think through if there is direct discrimination. Then you think about indirect effects. It would be quite helpful if officials started from the point of view of presuming that there is something going on that they do not like and then try to work out why that is not true, and therefore we can all relax and it is all okay. Going for each of those criteria, you take a region of the UK. Let’s presume that regions where we want to see more growth or where there are more vulnerable households will be the ones that suffer. Is there any reason why that might be the case? Once we can convince ourselves that, no, this policy will land well in those areas and there is no reason to think a particular policy would land badly, we can then move on. It is starting with that mindset of assuming that there is a problem in each case and working through it, reassuring ourselves that, no, we are wrong, it is all okay, that is great. Sometimes there will be data in the administrative data; sometimes those administrative data cannot support or help you very much, in which case the Department needs to be creative in thinking about either stepping back and using common sense or in some cases thinking what proxy data we can get. It is clearly hard to measure some of the characteristics that we care about in the administrative data the Government have access to. That mindset alongside when you are developing the policy can help. If you take a recent set of regulations that we looked at, which is where the Government are making it easier for people who are coming from certain humanitarian crises to claim benefits once they arrive here, those policies are clearly designed to benefit a particular group of people. You start to think through why they might have an unequal impact. You probably start to worry about whether the policy will work as well for people who are arriving here who perhaps do not speak English very well, for example. That starts to inform your communication strategy, because you need to get that right to make sure that work coaches and others administering the system can get the message across, can make sure that the right documentation and not more than is needed as being required so that the policy is not inadvertently not benefiting people in that situation.

One other thing that I will add is a very clear articulation up front of the base case, like the reference situation against which you are measuring impact. It often can be vague and ambiguous, particularly if we have a situation where there are multiple pieces of regulation floating around on a topic. What is the default here? We are measuring impact, but impact against what trajectory? Sometimes it is necessary to do two sets of impacts, because if you have two bits of regulations going on, you may need to be able to demonstrate the combined effect of the two. If you have done your thinking well at the beginning, that should flow out of it. For example, on the primary and secondary ones from the Way to Work regulations a while ago, where the stated intent was an increase in employment, that is the first impact, the first-order impact. Therefore, what we should be seeing is a demonstration of to what extent can different groups or different protected characteristics expect to have different impacts on this. Is there a reason why certain ethnic groups would have more or less elasticity to a change in the conditionality regime and so forth? There are two types of response. The first is the conditionality regime and do they follow it? Secondly, does that conditionality regime result in them activating more and getting into work? The secondary effect is sanctions as a result of their response, but if we think that the primary intent of the regulation was to get people into work, the primary impact that we are doing the impact assessment on should be on employment. The secondary impact, still important, should be on the sanction, working through the logic of, “This is a piece of regulation the aim of which is to get people into work. We are going to change the way that we are engaging with claimants. As a result, there will be different impacts—primary work. Secondarily, as a result of their reaction to the way we ask to engage with them, there may well be sanctions.” Again, none of that logic was laid out. It was just an immediate thing of, “What we can measure are the sanctions,” as opposed to saying, “In reality the first thing we should be talking about is who will this help get into work and why.” It is getting the theory of change. I have talked to the chief analyst about this and I am very supportive of her efforts. I have offered to help and I have offered Carl’s time as well to help talk through the theory of change and how to think about it so that the younger policymakers coming in have a stronger discipline about theories of change that they can then use as part of their policymaking process, which I think could be making a big difference.

Peter, I know you had an interest in the winter fuel payment question.

A few points on the winter fuel payments, building on what John just asked. What is your take on the regulations for 2024 and 2025 and your analysis that showed whether they demonstrated that they were delivering on the Government’s policy intent behind the two separate regulations?

Can you unpack that question a bit more? I just want to be sure that I am answering you properly.

There were two sets of regulations in 2024 and 2025 and behind those would be a policy intent slightly different from the Government from 2024 to 2025. Did you look at whether they were delivering on those policy intents, which obviously shifted between 2024 and 2025?

That is part of why we said the impact assessments were short of what they needed, because we did not see an adequate golden thread from a clearly stated—and I repeat that, clearly stated—policy intent through to a set of regulations, back to our point earlier about operation delivery, and the expected impact. It is easy enough, as briefed members of the committee, to understand in essence what the Government are trying to do, but we have always been very clear that we need that critical, clearly articulated policy intent. What we want exactly is, “What are you really trying to do?” We will never question that. Our job is not to question the policy intent but we will require and persist in getting clarity there, because against that we then do our work. In both cases the restriction and then the un-restriction the policy intent itself was not as clear as it could or should have been to be able to demonstrate—because if you have a loosely-worded policy intent, it becomes a lot more difficult to demonstrate if we have delivered against it. For example, on the second set, the use of the tax system as a way of clawing back, it was hard to see how that exactly matched what we were told the intent was because there were so many edge cases that did not quite seem to fit in. If the policy intent was that you want to tax winter fuel payments, we could have gone home after about half an hour, because that is what you were doing. But if the policy intent is to reduce the means-testing pressure of the pension credit limit and ease it, we will evaluate the way in which you have eased that pressure. From our point of view, the policy of the tax was a pragmatic, heuristic way of approaching something that was roughly right, but for us there were too many edge cases to say that that was the best way to deliver your policy intent.

That also threw up a set of questions, as Bruce said earlier, that, had we had more time, we would have gone out to consult on—open questions like using individual taxable income as your measure of whether someone should have to have their winter fuel payment clawed back or not. We felt questions like what about people with rent or disabilities that might impose costs or care costs, to what extent do they mean that? Taxable income is not a good reflection of the living standards of the household. Had we had a bit of time to go and consult, we might have found that some of those issues perhaps are far less important than we were worrying about and others might have been more important. There might have been new ones that were brought up, but they were the kind of questions that needed to be asked and needed to be answered if you were going to do this set of regulations really well.

You have touched on this a little bit. What lessons could be learnt for the future, particularly with winter fuel payments, around cross-departmental objectives to ensure that policymaking is more joined-up? Do you have a view on that?

There are perhaps different lessons to be learnt there. We have touched on some of the legislation around the emergency situations with Ukraine, Afghanistan and so on, where the DWP officials have done a very good job now of cross-departmental working to get more structural regulations in place. I do not think that DWP in particular is not well-versed in what it needs to do good cross-departmental working. We see that in housing as well. The issue is when it feels under time pressure and is putting solutions together quickly that are seen as pragmatic. That is where the challenges occur. We saw, funnily enough, less so with the first set of winter fuel payments but more so with the second set, that the Treasury or the HMRC side of that implementation was not as mature as the DWP side. It was not obvious that the means-testing paradigm, which has been well-developed within DWP about how to do it, how to think about it—Carl has mentioned some of those things—was translating over into how HMRC were then using tax as a form of means-testing. There is a big difference between an approach to means-testing and an approach to taxing. In that situation, by bolting on tax as a way of means-testing, a lot more work was needed to make sure that was joined-up if tax was the right way of doing it. I think that that was probably a case of rush and time pressure rather than an example of DWP not working well with other Departments, because when DWP has the time, we have seen it work well and sweat the details with other Departments.

Moving on to a different subject, the administrative earnings threshold. In March 2024 there was a report on the administrative earnings threshold, where you raised concerns relating to the evidence that was used by the Department to support further increasing the threshold. Have you encountered that type of challenge before and what was the reason behind it?

This is an area of policy where it is not easy for the Department because there is not a whole load of international evidence on how best to engage with people in low-paid work to get them to earn more. The evidence base is naturally thin. Our concerns were from, first, issues around policy intent from the start. Is this a policy about getting more people into work or is this a policy about reassuring taxpayers that their money is being used responsibly and that we are engaging effectively with clients, or is it a mix of the two? When we saw successive increases in the AET, we were disappointed that there was insufficient learning from previous experience of what had happened, despite it having been promised to us at an earlier stage that we would get an update and be presented with how it was going and what effects were being found. When we were presented with some pretty good evidence, there was not enough acceptance of the fact the evidence that we were presented with related to single people, which is an important group but they are not the only people who are affected by this policy. I think that when you think it through it is pretty clear that the further you increase the administrative earnings threshold, you probably expect the effects to get smaller. You do not expect them to be the same. We had a number of concerns with the evidence base, how that was being used but also the commitment to construct more evidence and learn from the process. For example, was this not an opportunity to perhaps try the AET in some parts of the country or with some groups to learn about who it works better for, who it works less well for? The AET was basically being increased in a relatively blunt way across the country. There was not, for example, consideration of the fact that younger people who have a lower minimum wage would have to work more hours a week to escape it. We did not see much thinking about people who might be struggling with health conditions but who do not have a WCA and how this policy will work for them and would work coaches engage appropriately with them.

Why would that not have been considered? Why was there that gap?

It is a difficult area of policymaking. There is not a lot of evidence out there and often the Government are trying to work quickly and trying to say, “We want more people into work, what can we do? We have a belief, a theory of change, that says that if work coaches engage with more people, that will help more people get into better jobs.” That is a reasonable theory of change to have. What we need to see is a bit more preparedness to experiment with different types of interventions—and you can see this in other areas too—and also a bit of patience. Saying, “We don’t quite know what is going to work, we are prepared to try some things, perhaps some of those things will not work, and we will wait a while to get the evidence before we go ahead” is sometimes difficult to do politically, but that is what we need to see more of.

This is a 10-year policy innovation. If you go back, in 2023 we were talking about it, 2013 the start of universal credit, the idea of in-work progression was being floated at that stage. Ideally, we would have been putting in place a 10-year policy R&D programme to say, “This is a genuinely new thing, we need to work out how it works, we need to gather evidence, we need to build an evidence base.” We have experimentation powers in the Welfare Reform Act that could have been used. The frustration at our end was that something that could and should have been evolved over time was treated very transactionally, “We are just going to change it. We are going to change it multiple times and hope that something will work, rather than building it up, changing it systematically over two, three, four goes, incrementally, and put all the testing, learning, the evidence base in to get it to work.” There was a combination of this being a wasted opportunity for something that seemed to have policy legs for people’s interest and desire to get it to work and at the same time it was a rush job to implement.

I think also that the Government were flip-flopping between saying that they were doing this to get more people back into work and doing it because people should not be on benefit and not try as much as they can to find work. It was expecting more of people getting UC, and so the fairness argument for the taxpayer was flip-flopping with the effectiveness argument of wanting more people to do more work, wanting to be more effective at increasing people’s earning capacity and the number of hours they work.

A policy objective can be a range of things, but both of those points could be in that objective for achieving both.

Yes, but I think that it started out with one and it moved to the other, rather than it being that actually both were—

You have moved me on to my next question about the focus of policy intent, and you just alluded to this. Essentially it went from getting people into work and better work into focusing on social fairness. Is that an unusual thing for the shift in policy intent in the DWP?

No, it is not unusual and it can be because there is a range of reasons why a Government might choose to do something. They might initially express it in one way, and if there is not a strong argument for that way, they move to another way of describing it. That is not uncommon.

It is not uncommon, but we are very clear that when we scrutinise, we will scrutinise on the basis of your stated policy intent. To your point, if the Government had put down two policy intents—we want to get people into work and we want to demonstrate fairness to taxpayers—we would evaluate it on the back of both of those. We may say that it clearly delivers the second, we have doubts that it delivers the first, “We would advise you to be pretty clear that this is predominantly about fairness to taxpayers, and we hope that it will have a benefit.” In effect, the position was much more the other way around. We saw that again last summer with the previous Government’s planned welfare changes that were, on the tin, about getting people into work, but were pretty clearly a cost-saving measure. The new Government took more of a cost-saving-measure approach to their policy intent and it meant that it was a more straightforward scrutiny job for us. When a Government say, “We want to save money, this is saving money” we will go through the process of checking to make sure there are no unintended consequences, but at least we are saying, “You want to do X, these regulations seem to do X, a few problems over here,” as opposed to, “You want to do X, but these regulations patently do Y.”

You touched on my next question about the impact on your ability to scrutinise effectively if there is a shifting policy intent from the Department. If it is not a one-off, it just happens as political messaging changes, how can we get over that? Is it a communication issue?

I do not think that there is a getting over it. As Bruce said, as I am sure you are all familiar, this is the reality of life in Westminster. Part of our job, and it is a very important part of our job, is to be holding up a mirror and saying that from the point of view of good governance the music has to stop, we all have to sit in our places and say, “That is what we are doing, and that is the reason why we are doing it,” and we will scrutinise on the basis of it. But if we get flip-flops of policy intent between various versions of documents, we will call it out. The power of the pen is pretty well all we have, but we are not shy about calling that out. What we hope to do is make sure that Committees like this and the secondary legislation scrutiny Committee, are well aware of our concerns and can amplify with their own voices as you and they see fit.

A final point on the AET. I believe that in your report on that you raised concerns about increased reliance on work coach discretions. Could you say a bit more about that?

Carl will come in in a second on this because it is an area that we have been interested in from the point of view of doing independent advice. The background that we have spotted over the years has been that as policies get layered upon others, there have been times when we, as a committee, have been concerned about how that will get implemented. We are often told that on the ground that would be work coach discretion. If you think about that for one layer, that is fine and makes sense, but then you start to realise that a huge amount of burden is being put on the work coaches and their ability to exercise discretion. We are in a position of thinking: do we understand the sheer latitude and the amount of discretion work coaches have? Do they have the time to exercise that discretion? Do they have the information available to them to have the judgment? This is not just to make a random decision. It needs to be a well-informed judgment based on understanding the goals of Government policy, understanding the rules and the discretions available and understanding the claimant’s situation and trying to use good judgment on the ground to marry all three of those. That is no mean feat in short-time interviews, so it is certainly something we are keen to look at. We have been keen; we have not yet been able to get that work off the ground. Carl, do you want to speak about that?

On the AET specifically, it comes down to you exposing a greater group of people who are in low-paid work to the requirement to regularly engage with a work coach and to be making greater efforts and show to be making greater efforts to be seeking increased earnings. One obvious question is might there be some people in that group who have some particular reason why it is more challenging for them to do that. How do you make sure that you are not inadvertently exposing them to sanctions when you do not wish that to be the case? If the answer comes back that work coaches can use their discretion, the obvious follow-up question is: are they given the clear guidance? Do they have the skills, the training, the time to be able to apply that consistently and right? I mentioned earlier the example of somebody who is struggling with a health condition but they do not have a WCA, they are doing a bit of part-time work, but they are now going to be caught by the rise in the AET. Are we confident that a work coach will treat them in the appropriate way and that equivalent individuals will be treated in the appropriate way in Scotland, Wales and Northern Ireland? How do we get reassurance that that is the case? As Stephen says, when it is multiple occasions of changes coming to us that rely on discretion being applied well, it raises that question. It is close to the issue that we started with about the policy to land well and to do what we want requires it to be operational and to work well on the ground. A key question and a key risk for the Department is that there is not enough attention in this area. Looking forwards, one can see that when we move to a world as the previous Government and the new Government intend, there will not be a WCA and work coaches will be engaging with a greater set of people on universal credit. They can be informed by whether the person is receiving PIP or not, but there is a difference between the PIP test and the WCA test. How can we be confident that work coach discretion will be used appropriately and consistently in that world?

It is not to say that work coach discretion is the problem. It might very well be the right answer. The question is more how a work coach will be enabled to make those very difficult judgments reliably and relatively consistently, because when you have discretion there will be variation.

My question is specifically on the independent research, but I follow up on Peter’s excellent questions and your equally excellent responses about evidence. As a former academic, I believe that evidence is absolutely evidence. You gave a couple of examples where the lack of application of evidence was a concern, particularly if we are thinking about quality policymaking. I appreciate that politics are added to that, but is it the view that evidence is not always applied, that it is more selected rather than having a comprehensive understanding? Are the different types of evidence understood and used appropriately?

That is part of it, but the thing I would stress is in areas that are really important where there is not enough evidence, can the Department not do more to say that in 24 months’ time this will still be a really important policy issue, regardless of how successful policy is between now and then? Therefore, we need to think now how we make sure that in 24 months, in 48 months’ time, we have better evidence in this area? If you take, for example, the crucial issue of the number of working-age people who are flowing on to ill health and disability-related benefits, it is pretty disappointing, given that that started around the summer of 2021, that we still do not have a good understanding of why it is occurring, let alone an evidence base that is starting to build up around what works in helping those households with their health conditions, helping them manage them and helping to get people into work, get people into a happier, healthier space and reduce the onflow to those benefits. In the summer of 2022, you might be quite sympathetic that this only started a year ago and we do not have the evidence. It starts to get a bit more tiresome that we are still flying a bit blind. The appetite to generate new evidence is the thing I would stress.

Public health academic colleagues might disagree with you not knowing about that, especially at an international level. Stephen, you were going to say something.

One often hears the phrase “evidence-led policymaking” and then the reverse of it is often taken as a joke, “policy-led evidence-making”, but I think that there is a point to this. There is quite a big industry of research papers that come through and are funded and supported by DWP, but that seems to sit in one part of the Department. What we do not seem to have is adequate join-up between the policy teams and that research. We do not seem to have as much policy-driven evidence-making in what are the policy questions and hence what evidence we need now and in two years’ time to support ongoing policymaking. About a year ago the Treasury produced its list of the research areas that it was focusing on. It had a very large chunk of that document devoted to what would be naturally Department for Work and Pensions research areas. That raises two questions. First, why is the Treasury doing it? But a much more important one is why does the DWP not already have that list, own it and is driving that agenda? That is the issue. We have advised the previous Secretary of State, Mel Stride, and Liz Kendall, when she was Secretary of State, and Stephen Timms, on the research agenda that we identified as necessary to underpin major disability welfare reform. This was not to say that we did not think that the Department had all this, but we wanted to lay out, if you are going to embark on what you already plan to be a multi-year transformational exercise, this is the corpus of knowledge that you should either have or be gathering to be properly informed as you go through that. As I said, I expect that they will have a good chunk of it, but I do not think that it is brigaded and I do not think that Ministers and senior policy officials have access to all that in an easy way.

I am sorry, Carl, that was not meant to be dismissive of what you were saying. For example, talking about NEETs, there is quite a good evidence base, at an international level and recently with the UK millennium cohort on NEETs and child poverty and the relationship there. Stephen, policy-based evidence suggests a selective use of evidence.

I was meaning to flip a phrase, policy-driven evidence gathering, if you see what I mean. I have this idea about where I want to focus. What evidence do I need to inform that decision? I am absolutely not suggesting any cherry-picking.

You mentioned conditionality and sanctions. The evidence base is so strong in this area on getting into work. Do you think that sanctions do not achieve people getting into work and do you think that is understood now?

In parts. What is missing is the flipside, which is a consolidated, synthesised view of the type of support that is needed to get people into work. What we have is the iconic piece of work that was done 20-something years ago out of the Treasury about jobcentre appearance frequency. If they turn up more frequently, they get into work faster. That was true and accurate in a time and place for a particular cohort of people, but that idea has, in effect, permeated the entire structure of the welfare-to-work programme within DWP. We are now dealing with very different cohorts of people, a different labour market, and we have not replaced—and we have continuous debates, as you have just mentioned, about do sanctions work, does conditionality work, and so on. I use the word “consolidated” advisedly because there is lots of sporadic evidence around the country of different programmes. What is missing, and I know that people in DWP are working to build this up, is that positive theory of change of different cohorts needing different type of help, here is what is effective, here is what is not effective, and hence let’s build a model around this knowledge, as opposed to clinging on to only one data point and the entire model being built around a frequency of interaction in the jobcentre.

Is there an understanding about systematic reviews and meta-analysis and so on that is there?

An understanding, but I would say not—I think that we, as a committee, would like to see that being much more systemic within the Department.

It is always open for Ministers to be sceptical about—I used to be an official—what people like me tell them is the evidence base. You cannot take the politics and the political analysis out of the interpretation.

Absolutely, but you need to have it. I will move on now to the independent research. Can you tell me what you are working on now and how you determine what you will be trying to commission externally?

On big pieces of work, we think carefully that there might be areas that we feel we want to get more on top of because we have seen a set of regulations in a particular space or we know a set of regulations are likely to come. That is helping to support our ability to scrutinise those regulations. A long-standing piece that we have wanted to do but have not yet got going on is on work coach discretion, because we see that as an increasing issue. We are open to people suggesting to us things that we might want to look at and we sometimes have had external people come in to present on issues that they think we should be shining a light on. It is also okay for Government Ministers to come and talk to us about what things they would particularly like us to look at. When we are deciding what to look at, we are very conscious of the fact that we are a very small group with very limited resources, so we cannot replicate what the Department does on its research, or an external body. We are very much thinking about why would SSAC be a good organisation to do this; what is it about this area that means that SSAC should take a look? Often it is because we are independent of Government and yet we have access to the Department and working with the Department means, for example, we can go and talk to some of the staff who work there, which often externally you cannot do. That is one of the key things that makes us determine, yes, this is why it is a good piece of work, not just for somebody to do but for us to do. Then from time to time it can be because we think something has been neglected. Stephen mentioned the two contributory benefits for out-of-work working age people, which we felt had not been looked at in a very long while. We felt it was time to look at how well they are working alongside universal credit. A piece of work we have going on at the moment, which Stephen also mentioned, is thinking about 16 and 17-year-olds and the support their household gets from the benefit system and how that is influenced and affected by the choices they make about whether they stay in full-time education or whether they do an apprenticeship. Our view is that this is not something that has been thought about in a long while. Despite the fact that we used to have a compulsory school-leaving age of the summer after turning 16, now we have a compulsory education or training age of 18, and yet universal credit has remained the same in that situation. It is early days in this project as yet but, for example, the amount of support a household gets can change really quite dramatically. If you took a single mother living with a 16-year-old; she is in paid work and she is on universal credit. If that 16-year-old chooses to do an apprenticeship, she may well find that she loses child benefit, loses the child element of universal credit and can lose a work allowance. She may also, I believe—and again this needs to be checked more thoroughly—lose access to some child maintenance support. The effect on her personal income from her child’s decision on whether to stay in full-time education or to do an apprenticeship can really change. We want to document that, work out the magnitudes, and then ask the question: are we happy with this? Have we got the incentives right? Do people understand this support? Is at least the young person making a decision that is in their long-run interest and understanding how it will affect their household in the near term? Are there improvements there? We are looking at other cases like disabled young people, young people with caring commitments, various circumstances.

Very good. Do you have any intention to do something on what was announced by the Government recently about the youth guarantee? Do you know if there are any regulations that may be associated with that that you will want to look at?

If there are regulations that flow from that, we will definitely want to look at them. It is important policy context for the study we are doing at the moment. The other thing, which we have got a little bit lucky with, to be honest, is of course the Department now has responsibility for more of the skills work and apprenticeships, which wasn’t the case when we started this in the summer. The Department should be even more interested in what we say in this report than it would have been. I think it is also fair to say the Department has been relatively encouraging on this piece of work in getting us access and wanting to hear about what we are doing.

Thank you so much for that. I want to follow up now on disability benefit reform. You may be aware of the work that we did as a Committee around safeguarding vulnerable claimants. I have got a quote here that you have said, “History has shown that reforming health and disability benefits is fraught with risks.” Were you thinking about, for example, the implications for our most vulnerable claimants when you made that comment, and particularly the impact that that might have on their health and wellbeing?

On the back of how they might respond to that, we were also thinking it was more risks to the policymaker and the policymaker’s expectations, whereby changes to the levels of payment are relatively predictable but changes to conditionality, thresholds, the tests and so on, which have often been designed at various stages to change the profile of the onflow on to those benefits, rarely end up having the predicted effects predominantly. Carl, you can probably speak to this as well given your research. Many people have other as yet unrevealed issues that were unnecessary to reveal to get however many points they needed for PIP, but if they need more points, other issues will come into play. The IFS has studied some of this over time.

I think that the transition from DLA to PIP is the classic example of that, where the clear, stated intent of the policy change was to reduce the number of claimants by 20% and the amount spent by 20% by changing the profile of people who are able to flow on. The Office of Budget Responsibility produced evidence on this. Its take is that the reform ended up costing money, so slightly more people were eligible, not fewer, than if we had left the old system in place. That is a risk for a policymaker because you don’t end up with the number of claimants, perhaps the type of claimants and certainly not the spending that you had hoped for. The whole process is also then you have carried out a reform where you have been engaging with people on the idea of saving money and reducing eligibility, and perhaps you are missing an opportunity to engage with a set of people around how to better design the existing budget. If you had started from the outset saying that we want to spend roughly the same, perhaps you would have engaged in a different way and ended up in a better place.

Yes, or understood the cohort and the change in population health status might have also helped.

The PIP change, to go back to the original question, not only failed to achieve its objectives but it actually damaged people, which, in a sense, makes it doubly bad. That, in turn, makes it also difficult for successive Governments to achieve its stated objectives, because some of that damage gets turned into a lack of trust. It gets turned into fear. You mentioned earlier the Right to Try Work regulations. In part, that is there because people do not trust the Department when it says, “You can go into work and it won’t affect your benefit.” You have to then do other things to try to build up that trust.

Absolutely.

As an ex-bureaucrat, I get very excited about things being ineffective as well as counterproductive.

Thank you. A final bit from me before I hand over to John. You have queried plans to link the UC health element to those in receipt of PIP. The two benefits have historically proved different types as you have mentioned a little bit already. The Government have reasoned that the purpose is to decouple access to UC health from a person’s capacity to work. Does this explanation make sense to you?

I think it is a clear enough motivation for what the Government want to do and, indeed, it follows on from a report that we wrote a couple of years ago on that very topic. The idea of saying that we want to make sure that the choice of going into work is not inhibited by risk of suddenly getting reassessed makes complete sense. There is also a question of how do you implement it? This is where I think the Government’s choice of how to do it introduces a level of cognitive dissonance into the system, which we think is unhelpful. Let me explain. The history of these two benefits is quite different and the stated purpose of the two has historically been quite different. To state the fairly well understood distinction, PIP is a benefit that is there to support those who, as a result of their disability, have additional costs of living. It has been a cost of living support and it has been historically independent of work status or earning, “It seems that you, as a member of society, because of your situation, have extra costs of living and we are going to compensate you for that.” There are plenty of debates about the adequacy and so on, but that is not the point. The principle is extra costs, compensation. On the other side, we have had, through a history of different benefits through the ages, a system of saying because of certain limitations—and they may not be the same limitations or reasons to have extra cost of living—we have, as a society, limited or no expectations of you being able to earn paid work. As a result, we will provide you with a higher level of long-term benefit because of wage scars and so on. I don’t want to go into all the details but you have, in effect, for good or for ill, two very different purposes behind these benefits, two very different reasons to test people, and two historically quite different criteria behind the tests, albeit a significant overlap. Should the Government want to say “We want to have one determination of disability and incapacity to work, and so on, and from that we want a certain number of rights to flow,” that makes sense, but it needs to start at the top by saying, “What is the principle behind what we are trying to do here? At the moment, we have two principles, two purposes; they can sit side by side each other from a principle point of view but they have become operationally enmeshed and confused. We want to disentangle that.” Fine, but you don’t do it by simply saying one piggybacks off the other. You have to go up a layer and say, “What are we trying to do here? We have a group in society, some of whom have extra costs but are perfectly able to work, some of whom are very limited in their ability to work but may or may not have extra costs.” The costs and the ability to work—we need to find some common thread that allows us to start and then flow down to what we want to do. Without doing that, we will continually find ourselves regulating against each other and finding exceptions and tweaks here. There will be barnacles on the ship before we know it if we do not get the principles right at the start.

When the Timms review is published with the new PIP regulations and so on, will you be seeking to look at that?

It is something I have shared in letters to the Secretary of State, I have discussed in conversations with Minister Timms and will continue to do so.

It is not something where Ministers are required to consult us. This comes into the area of a good working relationship.

Yes, but there will be new regulations.

Yes, but as we discussed earlier, depending on the Government’s appetite for our input, if those regs come in within six months of being laid, they not have to come to us for scrutiny, so they could get bypassed. On the other hand, maintaining a good dialogue, we probably can have influence informally as much as through that scrutiny process.

Yes, lovely. I will hand over to John again.

It is a start, is what I would say. It is obviously not the only thing that DWP is doing. There are also other committees, like there is the Right to Try Work Collaboration Committee, the Disability Advisory Panel that has been set up, but a lot hinges on how they develop. The Disability Advisory Panel has a very broad remit of all health and disability policy. The relationship between that panel and the Timms PIP review is not clear and we have written to the Minister about this. One of our independent pieces of work was to look at how the Department involved disabled people and made a number of suggestions for how that could improve and we advised caution against the use of the term co-production. We used a very tight definition and, in fact, we were sceptical—

The definition we used is, “Are you working equally with the group you are co-producing with to produce conclusions that both sides will own?” Say I, as an official, would work very closely with a group of disabled people and we would come together with proposals. If they were in a policy area, that would then need to involve Ministers very closely so they were tied into the process. Officials, the disabled people and representatives of other organisations were tied into that. We were sceptical about whether it is actually possible to do it in that way in developing new rules for benefits because disabled people are not going to be in the room when the decisions are taken. They are not going to be there in the meetings between DWP Ministers and Treasury Ministers. They are not going to be in the Cabinet room when they are signed off. They will be working very closely, potentially, and helping to formulate the proposals but, ultimately, there is an area that they are excluded from. That does not mean you can’t have co-design, very close involvement, but we were sceptical that you could actually co-produce in that very equal way but also benefit. You can definitely do it when it comes to the operational details of something, you can definitely do it if you are framing communications, but at that top level it is harder to do. We said also that the Department should set out the parameters within which it wanted co-working to operate. The terms of reference for the Disability Advisory Panel and the PIP review are fairly clear about this. It will work to produce proposals that will go to the Secretary of State and certain things are off limits. For PIP, what is off limits is that there must be a four point score on one daily living activity. We also said—and I looked at this last night—that when red lines are set, it is still possible to have a productive working relationship on the detail of a policy but we advised the Government against badging that as co-production unless its partners agree. I might be getting over-fixated on the term co-production. The crucial thing is that the Department and Ministers are able to develop an effective working relationship with a group of disabled people, representing different viewpoints, so a plurality. It is possible to do that as long as you are very clear what the parameters are. I personally would advise against using the terminology of co-production loosely.

Yes. We also said that they might want to use that term but they should define what it means so that everybody who takes part know what they are engaged in. We said the Department set out its stall. It should define these terms, it should define what its ground rules are. If that is the case that is fine but it may not be the co-production that many people out there in the disabled community would expect.

I think it is better to manage a degree of change of operating model that can actually deliver success rather than badging and setting expectations that it will be completely different and not get through the more formal gateways later in the process. I would be saying let’s involve the groups as much as is practical, let’s badge it appropriately. We may be able to go further the next time, but better to get something that has as much input as possible and gets through to the other end than claiming co-production, having lots of meetings where people think everything is great and then decisions get taken behind closed doors because this process didn’t deliver something that could be decided upon later. In a world where you are changing an operating model, you do not want to have a massive change to start with because you risk failure further down the steps.

Yes, it is. That is why, in general, when the Department has been in this situation before it has said, “This is the envelope we are working within; please help us to develop the details within that.” That happened to some extent with PIP, although it is arguable that they were not sufficiently clear that this was a cost-saving measure even though everybody knew it was. The reality is that the Department finds itself in a position where it needs to save money in this area. It can’t allow costs to carry on increasing in the way they have done. It can’t do this effectively unless it is working with disabled people to define some of the rules and turn its framework into something that will work on the ground. It is extremely hard to do. Obviously, they cannot expect disabled people to say, “I want to be paid less.” Social security is different in that respect from on, say, social care or health care, where you can say, “Well, what is the most cost-effective way of enabling someone to get on with their lives?” You could have a productive discussion around what is the best way of doing that within a certain financial envelope but you can’t do that with social security because the envelope is money, it is not service.

Before we move on, Bruce, I suggest that you have a look at the report that we did on the Pathways to Work programme. We challenged some of the points that have been made about the escalating working age benefits as a percentage of GDP, which we were advised from academics is the best way to compare. There is an increase but the ballooning metaphor really—well, it is not even a metaphor—is unhelpful. Also, we are about to publish a letter from Sir Stephen about the terms of reference for the co-production but I take on board the points that you made there. The last question is from Peter Bedford.

Turning to contributory-based benefits, what are your initial views on the proposal set out in the Pathways to Work Green Paper around the future of contributory benefits? How far do they go to address the challenges that were set out in the 2022 paper on that topic?

The first thing to say is it is great that we are actually thinking about these benefits and that they are no longer being neglected. That has to be a good thing. The Government has been clear about what the weekly rate will be. It will level up so that those who would have been on new-style JSA will get the same rate as those who would have been on the support element within the new-style ESA. There is more money per week in the initial weeks of receipt but the benefits will be time limited for everyone. We do not know what that time limit is; that is yet to be determined. I think it is quite plausible that this reform will save money because at the moment a lot of the spending goes on people who have been on new-style ESA for considerable periods and they will lose out from the reform. What I really want to see—and I know this is still being worked out by the Government—is thinking about all the other aspects of these benefits. The self-employed will be brought into new-style JSA, effectively the new benefit, for the first time. How will we measure whether somebody who has been self-employed has built up a sufficient contributory record? At the same time, we should not miss the opportunity to think about how we record whether somebody who has been an employee has built up a sufficient contributory record. One of the things we highlighted in our report is that the measure is quite odd in some cases. It is about the amount you have contributed in two recent financial years. It can be the case that you could apply in December of one calendar year and find you are not eligible but if you waited until January, because of the calendar years they would look at, you would actually become eligible. You would think that through using PAYE data we can do something a little bit better than just taking two years of snapshots. We could perhaps think about a longer period. We could perhaps take into account more recent weeks rather than looking back to a financial year that may be eight or nine months ago. I think that they should also look at the payment frequency. In particular, when we thought about people on new-style JSA, they look to us like a group who could handle monthly payments pretty well. Paying them fortnightly if they are also receiving universal credit can be pretty odd because universal credit would much rather people receive money monthly rather than a different frequency, so I think there is an opportunity there. Another thing that struck us when we looked at the current new-style benefits is that there is a means test for those who have individual pension income, which I think looks increasingly odd in a world of defined contribution pensions and pension freedoms. That means test should be looked at again in the new world. The overall message is that I am very pleased they’re doing this and I think they should look at lots of the details, not just fixate on the amount and the duration for these other details. It is an opportunity to get them right and get them working in the context of universal credit, the modern pension system and other changes we have seen in the economy.

You have answered most of my next question, which was basically going to be what are the details you would like to see in the White Paper when it is published? You have listed quite a few things you would like to see clarification on. Is there anything else you would like to pull out that would be good to see once that is published?

Starting with clear thinking about what this benefit is for is always a good place to start from. We have talked about that a lot today. Then I think getting all of those details right. Lots of the other details just have not been updated in a very long while, despite Government having access to better data, despite policy changes in other areas, despite the fact we have introduced universal credit. I think all of those changes mean that what we currently have in place is not working as well as it should do.

One particular area of planning that we identified in that report, and actually have subsequently been looking at, is the RTI system where data sharing, even within Government let alone what employers are feeding in from the two different benefits, could be done a lot better. To Carl’s point, that could be used as having a basis for a very different and more sensible way of determining national insurance credits, since we have, in effect, real-time data for the best part of 10 years. There is no reason why we have to have complicated eligibility rules, rather than simply taking on principle X amount over the last X years, simply captured, would allow people to have much greater clarity. There is also making sure that the data about the payment of one benefit is automatically fed into UC so that those calculations are done. RTI has been an area that we have been looking at. This is perhaps a second order issue but it is a time to get a lot of this sorted out, given that a new system will have to be built for them.

I have another one to think about. Another recommendation we made was that when employment support schemes are brought in, the default should be that they should also apply to those on contributory benefits. We thought a good starting place is that those on contributory benefits should be getting at least as good a deal as those who have not paid into the system. That does not mean in some cases you might say, “Well, no, this policy is better designed for the UC workload, it is less well suited to the contributory workload” but we thought the default should be that that support should always be there. We were pretty struck by hearing examples of the IT around the contributory benefits just not being up to date. People basically described how they go in and talk to their work coach, they are handed a piece of paper with when their next meeting is and they would much rather have a text. They do not get the equivalent of the UC journal to engage with. As I say, these are often people who have been in paid work relatively recently. They have mobile phones and are relatively well placed to engage with a UC journal. If you were going to start anywhere, you would probably start with this group. There were some really disappointing cases where people, if they are on both benefits, told their new-style work coach that they had found work but they had not told their UC work coach, who is somebody different, and the system wasn’t talking internally. Again, that is an example where, from the claimant’s point of view, they have told the Department; from the Department’s point of view, they have told only one part of the Department. There are examples of where the system as a whole can work better for people.

Thank you very much. That concludes our session today.