Good afternoon and welcome to the Women and Equalities Select Committee. Today we are continuing our oral evidence session on the effectiveness of equality, diversity and inclusion initiatives, which was previously disrupted due to votes—many, many votes. Today we will hear again from Dr Louise Ashley, reader and associate professor at Queen Mary university; Peter Cheese, chief executive of the Chartered Institute of Personnel and Development; Iain Mansfield, director of research and head of science and education at Policy Exchange; Dr Zoe Young, founder and director at Half the Sky; and, online, Ama Ocansey, UK head of diversity and inclusion at BNP Paribas. Welcome. Can you hear us, Ama?

I can hear you very clearly. Thank you.

Thank you very much for joining us. I am going to hand straight over to Kevin.

I would like to start by looking at the Inclusion at Work panel and the Government’s roles and responsibilities in assessing what works. Starting with Zoe and Peter, how can employers establish a proportionate, cost-effective approach to collecting and analysing the workforce data needed to measure diversity and inclusion in their organisations?

Much workforce data already exists, as many organisations have systems for holding information about the people they employ. The Inclusion at Work panel observed, and I see this through my work as well, that it is more common to use static measures, such as the make-up of an organisation and where people sit within it, than dynamic measures of progress and process—recruitment and churn, who progresses and how far and how fast. The data exists, but it is not always interrogated in a way that helps pinpoint where to focus intervention, so it would be useful to have specificity and guidance around that. Data can also be qualitative, yet there is a reliance on quantitative data that reflects the focus on representation. There is also a bit of an understanding and capability gap around the fact that qualitative data can be good science and can help demonstrate and explain why certain initiatives work or do not for certain populations.

I would add that, while organisations are collecting data, it is pretty inconsistent. I echo Zoe’s thought that, if you are collecting data, what are you doing with it? It is not just about representational statistics, although those can help give you a picture of your organisation. I would absolutely look at things like progression and retention. Particularly important is this more qualitative understanding of the organisation’s culture. Is it supportive and inclusive? Do people feel that their lived experience is fair and that they are treated fairly? It is these cultural dynamics, for want of a better way of describing them, which are absolutely elemental to creating positive workplace cultures that support the idea of inclusion. The more qualitative data can be picked up through the fairly classic channels, such as employee service, but they can also be picked up from things like exit interviews. While there is information out there, what the inclusion panel shows, and what we continue to talk about, is that organisations could almost certainly do with more guidance on what are appropriate measures and how to use that information to support good practices around inclusion, and to support and train line managers to help them understand their role in supporting inclusion and positive workforce cultures.

As we are talking about a proportionate response—for businesses as well—qualitative data can take more digging into. Does that put a higher burden on teams that are analysing it? Is it more difficult to gather? Is that an issue?

Qualitative data, like quantitative data, has sampling purposes and thresholds and validity. It is skilled work, just as quantitative data collection and analysis is, but I think that what is not well understood is how much qualitative data you need to generate before you can draw a conclusion, and the evidential value of qualitative data from interviews or group discussions, for example.

There are lots of tools and methods for capturing this kind of data. Some of the quantitative data can be more challenging. When you look across different aspects of diversity—for example, disabilities—we know that in many organisations there is a challenge with people declaring, particularly when those disabilities are less visible, such as in areas of neurodiversity. But if people are not prepared to declare, that, in turn, can be an indicator: do they feel it is safe to talk about these things? Do they think they are going to be judged as a result? Certainly having more consistent guidance around what measures make sense and how we understand both the qualitative and quantitative dynamics of an organisation would be helpful.

The CIPD has said there is an appetite among employers for more standardised metrics and more guidance from the CIPD and others on this. What progress has been made on that?

We have certainly done a lot of work, talking to many organisations about the things they record and report, to try to come up with more consistent standards. There are other organisations that can support this. For example, the British Standards Institution has done work on international standards. It can even link to what organisations report externally—I am sure we will get on to that later across the panel. So there has been progress, but there is always that question of where these things come from. Standards bodies, or professional bodies like ours, can very much help to move things in a direction, which we have certainly been trying to do, but as I said, our data continues to show that a lot of organisations are not reporting on these things consistently. However, that is with due deference to the recognition that, as Zoe said, this is not always easy for many smaller businesses. I have always said about collecting this kind of data that if this was easy, we would have doubtless done it years ago. For smaller organisations, which make up the majority of organisations across this country and most others, it can be a little harder. So we need to be careful that what we are providing guidance on is practical and makes a difference, but is also simple and does not get overly complex.

What do you think the role of Government should be in this? Are you aware of any work the Cabinet Office has been doing around this?

There are always two or three roles for Government. One could be, what would you require organisations to report on? Obviously discussions about ethnicity pay gaps and so forth will be part of that. But I also strongly believe that part of it is about the guidance—and perhaps even direct support in some cases—that Government can provide. That might come through other arm’s length bodies: ACAS, of course, is a good example of that. The third one is the Government’s convening power to bring together different bodies—standards bodies and the like—and say that maybe this is the time to push this a little harder so we have more consistency. It is across those three areas.

There is a role for Government in bringing to life the idea of the Education Endowment Foundation toolkit-type assessment model; something practical and usable that will help decision makers compare options for what they do on diversity and inclusion, based on the strength of evidence, the likely impact and cost. As we discussed when we last met, diversity and inclusion interventions solve different but related problems. The first requirement of any such assessment model must be clarity about the problem it is addressing, be it representation, disparities in pay or progression, workplace safety, team performance or behaviour. A second requirement must be the quality of evidence, and in many areas the evidence base is still developing, and that should be visible. The third requirement is about context. Workplaces vary significantly across sectors and job types. The conditions in a hospital, in a technology firm or in retail are very different, and that shapes whether and how diversity and inclusion interventions work. So I would argue that there needs to be a sectoral, contextual lens to show where interventions are likely to translate, what conditions matter, and where adaptation is needed. If I may offer an example, bystander training, which has been widely adopted, or speak-up training can be effective as an intervention on their own in teams where the hierarchy is quite flat, where people routinely work together, and that is because people have permission and opportunity to intervene in the moment. But in hierarchical environments and high-pressure environments, such as parts of healthcare and policing, decisions are time-critical and the authority gradients are steeper. To make bystander training effective in those environments it needs more system supports. Decision makers investing in interventions in this space need to understand that, in order to get behind the interventions most likely to be effective in their context. Without those supports things such as bystander training or speak-up training raise awareness, but they do not necessarily translate into what people do. There is a role in specifying that definition and measurement, and in bringing the evidence that you can rely on in your context to make that more accessible for employers.

That makes sense; I have worked in very hierarchical teams in healthcare, and it is very different to other types of employment. Ama, can I move to you? A key recommendation from the last Government’s Inclusion at Work panel was that the Government should develop an assessment model for EDI initiatives, similar to the Education Endowment Foundation’s model for assessing teaching interventions. Can you explain how the panel came to make that recommendation, and how it would work in practice?

We came to consider that recommendation because different organisations are at different stages of maturity. It was felt that a toolkit providing guidance on steps, wherever you were in your organisation in terms of maturity, could allow you to progress. For example, the disability at work programme has different tiers—level 1, level 2 and level 3. That allows you to assess where you are in your organisation in terms of disability. If you are a grade 1, you can then see how you can achieve grade 2; subsequently, if you are a grade 2, you can see how you can achieve grade 3. Because of the different methods and approaches to diversity and inclusion, it was felt that if a toolkit like that could be created, organisations could seek the relevant accreditation, according to what stage they were at in their maturity. The way that that would work is that there would be a grading system. Each level would be assigned actions and achievements that could be made. For example, if you were trying to achieve grade 1 accreditation in the toolkit, one of the requirements would be that you would have to start collecting your data and looking at the current actions within your organisations to see if they address certain things. Different levels of accreditation would ascribe different achievements. Obviously, the higher you got, the more you would be required to have a deeper level of maturity, a deeper level of tracking progress, and the ability to evidence how your interventions or activities would lead to progress and change. The purpose of the grading system is to make sure that people are not disheartened or put off. The trouble with diversity and inclusion is that a lot of organisations do not know where to start, and those that are implementing initiatives are not necessarily tracking or measuring their progress or impact. The ambition with this toolkit would be that these levels would allow you to start somewhere and give you strong guidance as to how to measure your progress. That would also allow an external organisation to review what you were doing in your organisation; obviously, if you ask people in an organisation what they are doing, you get very different levels of thought. But if you have an external organisation saying, “This is what you should achieve, and in order to have the accreditation, you should have done this,” that allows the external body to review you, but it also means it is very clear to you what you are trying to achieve and what you want to do.

We are looking across the whole employment sector, and this slightly speaks to what Zoe was saying about how different organisations can be in the way they operate and in their team structures and so on. Is there something about taking what was essentially an education model into all workplaces and employment situations? Is it more difficult to assess diversity inclusion across all these different organisations because there are just more variables across the board in different organisations than there are in the classrooms?

Absolutely. Different organisations have different levels of resources. If a small trader or a small accountancy firm is asked to implement diversity and inclusion initiatives, that will require resources—human resources, and tracking your recruitment and retention processes—and will incur additional costs that they may not necessarily be in a position to absorb or expend. If you are a big investment bank, that may be part of your HR management budget. The education model is good in the sense that it will allow you to explain and it will allow people to access that toolkit, but it definitely has to be understood that organisations are varied. By using this tiered model, like I say, you can meet organisations where they are at.

Alex Edmans’s view is that a Government-funded and developed assessment model might actually stifle diversity and inclusion innovation. Do you agree with that take, Ama? If so, how can we mitigate that risk?

I have two competing thoughts. If something is voluntary, it allows people who are really interested and who really want to make the change to engage. But the problem with being voluntary is that it allows people to easily opt out. That is one conflict that I am holding. The other is that if it becomes mandatory and compulsory, you get organisations that are not necessarily committed or interested, and they will do things in order to tick a box. This is the kind of space that I am in, but if I was actually pushed, I would say that making it mandatory is probably the better approach., It is cumbersome, there is more bureaucracy and it is more layered, but when I look at the gender pay gap reporting, however difficult that is, it does make organisations look at, analyse and explain their data. Even if you are an organisation that is not committed and you are just going through the process, you still have to look at that data, and by going through that process, things will change. It may be quicker in organisations that are committed than in those that are not, but I am very much leaning towards the view that making people do things allows them to engage, and hopefully the next steps will mean that they try to do something about it.

Louise, you were nodding. Is there anything you wanted to add on that at all?

No, I thought that was a really brilliant summary of the tensions and challenges around this agenda, and I would broadly come to the same conclusion on the mandatory element.

Thank you all for coming in again. Zoe, to what extent do you think EDI can be a drag on companies’ productivity?

Oh, my goodness! I thought about this question initially with this framed as the EDI sector, and what that is. The answer is that it is not a single, coherent sector, a cost centre in an organisation, or a function. It spans legal advice, data analysis, HR, organisational development, consultancy training and behaviour expertise. It is very difficult to break that down. Like any area, the impact and the utility of this activity depend on the quality and the focus. That is where we have had some variation in responding to organisation-specific challenges, and myriad interventions, some of which may be more about signalling than substance of change. I feel that diversity and improving organisations’ access to the widest possible talent is an engine of growth and productivity. Therefore, effort to remove barriers that prevent organisations from being able to select from that wide pool of talent—wherever it comes from and whatever human form it takes—is worth doing and is fully supportive of productivity.

Ama, I am quite interested in your views on this too. Being a large multinational, as BNP is, how do you assess the benefits and cost of investment in your EDI-related functions?

We look at data a lot, both quantitative and qualitative. The qualitative data is very much HR based: what sorts of people are being promoted? What sorts of people are being put on talent programmes? What types of people are leaving? What does our graduate pool look like? So the HR piece is very much the qualitative. We also look at quantitative data, but If I am honest, it is very much based around employees’ surveys, and there is fatigue within that world, so we do not always get the response rate that we would like. We have to use other things, such as anecdotal evidence around what our networks do, if they hold events or workshops, or have speakers in, to measure what employees are feeling. The thing about diversity and inclusion is that you have to have a diverse workforce, but you also have to have an inclusive workforce. Laid on to that, you have to have not only an inclusive workforce, but a workforce who feel that they belong. The diversity is quantitative, and that is easy to measure. The inclusion is qualitative, and the belonging is more nuanced. Apart from surveys and speaking to people, you have to almost try and assess your culture and encourage actions among leaders and line managers to cultivate that kind of culture. That is very difficult to measure. If you are really crude, you could say that you can look at attrition, but that is not necessarily an indication of a bad culture; it may just be that there are opportunities that people are seeking elsewhere, and just because people stay, it does not mean they do not find a culture toxic. What I am trying to say in a very long, convoluted way is that there are various measurements, but you have to be flexible and to understand that it is not a blunt instrument; you have to be quite nuanced, perceptive and selective as to how you respond and how you enact the activities that you want to progress.

That is really helpful. Just to go back to the original question, Ama, do you view EDI to any extent as a drag on your company’s productivity?

Sorry, I should have answered that directly. No, I do not. I think that it requires time and understanding, but as a global organisation, we have global clients and global employees. Your workforce is everything, whatever industry you are in, and regardless of the rise of AI, so you have to invest in your employees. I do not see how making people feel that they are represented and included, that their voice matters and that they are bringing their best selves to work can ever be a drag. Productive employees are productive organisations. What makes employees productive is being in cultures and environments where they feel they can thrive. How does that happen? You have to meet people where they are, make them feel included, make them feel they belong and respect them. I do not see how that can be considered a drag on resources. Your people are your main asset.

Thank you. Louise, do you want to add something?

Yes. I am a sociologist, so I will come in from a more sociological perspective. When we think about productivity, we tend to think of it from an economic point of view, and obviously that makes an awful lot of sense. We talked about this a little last time, when we talked about the business case, and I think we mentioned that one difficulty with the business case is that it may or may not exist, or it may exist in some circumstances but not in others, so it is highly contingent and a little uncertain. That is why it can be a useful driver for change, but slightly problematic as well. Sociologists have been very critical about the use of the business case as the primary driver for change in EDI. That is not to say it plays no role, but we need to recognise that contingency. That relates partly to its focus on economics and the degree to which it borrows from a classical economic model. I will not go over that again, because we talked about it last time, but the point I am trying to get to is that when we are talking about productivity from an economic point of view, it is important to bear in mind that we will never get a definitive answer. It is impossible; it is too complex, and it will be too varied depending on the particular context, what is happening and who we are talking about. When we talk about productivity, it will rarely be entirely evidence-based; it is always going to be politicised to some extent. This is partly because EDI is inherently political, although it serves some people to pretend that it is not. I am just putting a flag up that we need to use a little caution when we talk about productivity, because we will never be able to prove it one way or another definitively in every single circumstance, and we need to work with that as opposed to against it.

Iain, your report, “Putting Business Back in the Driving Seat: Countering the rise of EDI in the workplace”, argues that the size of the UK’s EDI sector is a “misallocation of labour that is costing the economy £10 billion”. Could you explain how you arrived at that £10 billion figure?

Could I just correct you slightly on that? We attempted to quantify the cost of EDI across a number of areas. The £10 billion is actually the disproportionate size of our HR sector compared to the US and the EU. We look at the cost of EDI across a range of other areas as well, including training costs, which we estimate at between half a billion and a billion. We look at the cost of individual proposed and existing regulations. For example, in financial services, the FSA and FCA estimated that a new piece of EDI regulation—although this was withdrawn because it had a good impact assessment, it is a good indication—would cost business around £312 million a year. There is obviously already EDI regulation in financial services, but that gives a particular proportion. We also have some major court cases. Some of these are individual court cases; we could look at Borg-Neal v. Lloyds, with an award of £800,000. There are various gender-critical court cases, which have cost employers over £100,000. Perhaps the biggest one, which gives us confidence that we are talking about in the billions, is an equal pay case against Asda, which will cost them up to £1.2 billion. There is a similar-sized claim against Birmingham city council, which is connected to the ongoing bin strike. We would contend that these laws are not operating in the way they were intended to operate, and that that is why there are these very heavy costs. This goes beyond making a happy, productive, inclusive workforce and is actually pushing against the market and pushing against productivity. To go back to your £10 billion on HR, it is an estimate. It is not all due to EDI, but the proportion of our workforce employed in HR is around 1.6%, compared to 1% in the US and 0.8% across the EU. We looked at three specific ONS classification codes: 1135, HR managers and directors; 3562, HR industrial and relations officers; and 4138, HR admin occupations. If you look at slightly different ones, you can get slightly different answers. We excluded, for example, vocational trainers, which some people would consider HR, but we did not. As I say, those are the ones we looked at to get the discrepancy. Between 2011 and 2023, we have seen an 85% rise in HR. It has fallen back slightly in 2024, but it goes from about 280,000 to around 480,000, which is a really significant rise. If you look at those ONS classification codes and compare them to similar ones—that is, professions that have the same first three digits but a different last digit—you see that HR is growing significantly faster than comparative roles, as well as the economy overall. We do not say that this is just due to EDI, and we are definitely not saying that HR is inflating itself or that companies are randomly employing HR professionals. We trace this back fundamentally to regulation flowing from regulators, which is driving this cost to business. Businesses are acting rationally—we are not saying that they are not—and that is where this cost comes up.

You are comparing the UK model, with UK ONS codes, to the US model, which is possibly different. Are you comparing like with like, or are you comparing apples with oranges?

We think we have done our best to compare like with like. There are always challenges going across different countries. We looked at some OECD data as well, but we think this is a fair comparison. Particularly when you look at the growth in the UK over these years, it gives us confidence that there is certainly some over-representation and over-growth here in the UK.

I do not want to labour the point, but I want to dig a little into your evidence around that £10 billion. You mentioned several court cases—including a pay dispute case with Asda—as costing those companies money and contributing to an EDI cost. Given that this is legislation based on the Equal Pay Act and that those cases enforce that legislation, if the company had employed effective EDI strategies and actually paid the women appropriately according to the Equal Pay Act, could it then have avoided those costs? I am just wondering whether it is the other way around and those costs have come out of not having effective EDI, rather than being an EDI cost.

We would say that, in those cases, the law is inappropriate because it is comparing two entirely different jobs—for example, warehouse staff and shop floor staff, or refuse collectors and those working in other areas of the council. If you dig into some of the cases, you find that economic need was not accepted as a reason for paying them differently and that, in fact, in some cases the companies had offered the employees in one to move to the other. You are also not looking at situations where you have massively dominated ones; at Next, for example, 75% of the consultants were female, 25% male, whereas in the warehouse it was 47% and 53%, so you had a large number of men and women in each. This £10 billion is not made by adding up these separate things.

No, I understand that.

What we are saying is that it is fundamentally the laws, including what we consider inappropriate laws. We specifically recommend changes to those provisions in the Equality Act 2010 that come into the cost of EDI.

Do you think there should be equal pay legislation?

We do think there should be equal pay legislation, but it should be tightened to refer to similar work or like work, not to work of equal value. Judges should not be comparing and assessing the value of work; that should be left to the market.

This is an important debate, and I think what concerns us is some of the headline messages that have come behind this. Yes, the profession has grown, but so have other professions across business. Business is more complex, with the legislation and a lot of the dynamics that business has been operating with over recent years. We do not think it is a reasonable, apples-for-apples comparison to compare two very different jurisdictions; it is too simplistic to make that the core argument about what the size of the profession should be here. We are operating in a different context. Also, when you look at other dynamics, yes, legislation has increased, but that then starts to impact smaller businesses as well. You can begin to understand why the profession has grown: there is increased scope of activity and certainly, we would argue, more penetration. We have been working with the ONS directly, not at the three-digit level, but at the four-digit level and the six-digit level, to better understand these numbers, because we would agree that the ways some of these things are counted are not terribly accurate or consistent. It is not just across jurisdictions; it is over time and what we are recording as activities within a particular function or profession. To my point about smaller businesses, they are also operating in these environments. In the past, they may have got away without necessarily having designated HR professionals, but they are much more likely to have them now. We have struggled a bit to understand where the argument is trying to drive from; is it driving from a legislation perspective, in that there is so much legislation that is not necessarily adding value? I do not think we necessarily need to get into that debate right now. But we do need that support. At the end of the day, businesses will decide what they need in terms of the HR function and profession, just as they do for other functions and professions. As Ama and others have described, it is a vital role in understanding the modern dynamic of the workplace to drive positive outcomes. My final point is that I do not think any of us is denying that the profession needs to be better at connecting inputs, outputs and outcomes. Whether it is EDI, wellbeing initiatives or any of these other things, what is the business rationale for doing these things? While I agree with Louise that we can try to over-calculate some of these things, it comes down to some of the principles that have been talked about: accessing all the skills I need in an organisation, having workforces that represent the communities of which I am part, having in mind the wider workforce I am trying to engage and support, and having a workforce that represents the customers they are serving. There is also lots of evidence that shows that this is about diversity of experience and background, which brings different thinking. We know that different thinking not only breaks the cycle of groupthink, which is quite important in many organisations, but drives innovation. So we have to be very careful not to make big assertions about numbers, which we have been investigating further and will be reporting on further, and which are not necessarily based on apples-for-apples comparisons. It is also about what we are trying to drive at: what outcome are we trying to achieve?

On Iain’s last point, not many people here will need to be reminded of this, but to go back to my point on politics, the idea that the market is a good way to decide wages and that it will do so on a rational basis is deeply flawed. It is a political idea with a lowercase “p”, which comes from classical economics—I do not mean party political in this case at all. One thing we know is that the market consistently undervalues work done by women. Therefore, if we leave it to the market to decide, we will continue to have unequal outcomes. Again, with the caveats that have been given, I am very wary of that notion.

Iain, I want to follow up on your rationale for comparing the UK and US HR sectors. I just wondered whether this was coming from who you are writing this for—an American audience or a UK audience—and the funding for that work. We know that around £5 million of your funding comes from the American Friends of Policy Exchange, and it has been reported that a huge part of that is ExxonMobil. Last year, ExxonMobil banned the flying of LGBTQ+ flags outside its building, its shareholders previously voted down transparency on social impact and pay gap reporting, and it has moved away from conversations around equality to meritocracy. I just wondered whether this piece of work was more for an American audience, or whether it was looking at this from an international perspective, rather than just being for a UK audience.

The two major comparisons we use are the US and the EU, because they are the major developed economies which make a natural comparison to the UK. As I said previously when you questioned on this, we are a registered charity; we are up to date with all our declarations in terms of our required submissions to the Charity Commission. Regarding funding from the US, certainly to my knowledge, none of the funding for this project came from there. To the best of my knowledge, we have not received any funding from ExxonMobil for at least five years.

To the best of your knowledge—that is great. Thank you very much.

When you look at the structure of an HR team, is it really possible to separate out the work done on EDI, compared to other essential HR functions, such as sickness absence, performance management or the various leave arrangements?

In general, it is not. That is one reason why, much as we would have liked to come up with one single number for “This is the cost of EDI in this report,” we did not. We came up with several separate numbers, which are presented separately, for different costs. We look at the cost of HR, which combines EDI and other functions. We look at EDI training. We look at court cases and so forth. I think the media ran with the biggest number, which it has a tendency to do. But that is why we thought the most objective and transparent way was to present a number of the costs across different areas of different sizes. Today we are focused on EDI. We have written on regulation in other contexts; we have written three or four projects in our re-engineering regulation programme over the last three years. We certainly do not think EDI is the only area where there has been too much regulation driving HR, but that was the focus of this report.

Just to clarify, your view is that it is not possible to split it out?

Certainly not robustly over the whole of the economy, or at least it would be extremely difficult to do so.

You can separate out people who are designated in specific EDI roles, but that is a very small number. We would be happy to share the research we have been doing on this with the Committee. The wider point that Iain makes, which we would agree with, is that a lot of good EDI practice is integral to most of the rest of HR practice. By their nature, good recruitment and good performance management should be inclusive. That is why it is difficult to separate these things out; this just becomes part of what managing a good business and what good people management look like, with all the different HR and talent management practices that go with that.

Ama, what is your point of view, with your business?

I would completely agree with that. I would argue that the number of EDI professionals is not what has inflated the HR budgets. In most organisations, the EDI team is very small. I would argue that inflated HR is due to increased regulation, and not necessarily EDI. It is also due to increased risk management and increased litigation. Business has become more complicated, and that is why it needs HR to manage those risks. To the last point that was made, good HR should include aspects of EDI, and those should be integral, but the challenge is that they are not. This is why EDI is necessary.

That is an important question. I would say not entirely. We have touched a bit on gender pay gap reporting, and our analysis has shown that too much tick-box reporting is going on today. For most organisations, relatively speaking, it is a lot easier to measure than ethnicity and particularly disability. What have we learned from gender pay gap reporting? It is encouraging that the Government are going to require a narrative. We have worked with the GEO, among others, to look at what is publicly reported about the gender pay gap, and many organisations just report a number, with no narrative. Interestingly, there is quite a spike when you look at zero, and the likelihood of a zero pay gap is extremely small. The wider point is that we have to learn the lessons from gender pay gap reporting. If we are going to make this have an impact, we have to explain the rationale. All these questions about compliance need to be driven by the question of why this would be important to business and what business outcome we are trying to achieve. We can certainly learn from the requirement to have narrative, as I said, and action plans that back up whatever numbers people are reporting. But it is more complex than that. There have been a number of reviews, particularly around ethnicity, race at work and things like that. The argument has been that pay gap reporting can encourage more ethnic diversity in organisations, while recognising that the pay gap is not just about the number of people, but about the fairness of pay and about progression—that is often where we start to see the pay gap emerge, because there is less progression, whether for women or people from an ethnic minority background. In other words, it is quite a crude instrument, and that is the challenge. There is also, as I have said, the complexity of reporting, even on ethnicity, which we can record in all sorts of different ways. And as I said a few minutes ago, when you get into disability, it gets much more challenging. Many people are not declaring, and there are a lot of so-called less visible or invisible disabilities. A lot of this speaks much more to the culture of an organisation and its ability to do these things. The final point is that if we are to progress on some of these things, we would suggest starting with large organisations. Obviously GPG reporting applies down to organisations of 250, but in many cases you are dealing with small numbers. If we are to proceed in this direction, we should maybe start with strong businesses of around 1,000 employees or something like that. But I would absolutely emphasise that we need to make it very clear—just as we do on any legislation around employment—why this would benefit businesses. Otherwise, they will increasingly look at these things as a burdensome compliance issue and will not understand what benefit it is driving for them.

It is to be proven, I guess, because we do not have it today. We have always encouraged, as good practice, that organisations have some narrative to explain things when they report these numbers. It is like any financial number: you would not have financial metrics without a narrative around them. You do not report your profit and loss and say, “Good luck with interpreting that.” You explain it. We have always encouraged that as good practice. To be honest, with mandatory action plans and things like that, it is still to be proven. But we have to make sure they are put into a context, so that even they do not just become about compliance, with somebody going on to ChatGPT and saying, “Give me five great actions to put in place” and boom—job done, box ticked.

To some extent, thinking about pay gaps and measurement relates to the specific problems we want to address in the labour market. There are issues around people within the same organisation who come from different ethnicities and who may be being paid systematically different amounts; that is obviously something we need to understand and tackle. Similarly with gender, although less so now perhaps, a lot of the pay gap comes through occupational segregation and the fact that the types of work done by certain types of people tend to be systematically undervalued. That takes us back to the question of women’s work, or work that is associated with women, including work in the care sector, and how value and status are attached to work. That is a very complex subject that goes far beyond what the market decides, but we know sociologically that the value attached to work is partly to do with its nature and complexity, but also partly attached to who does it. We know that work that is more likely to be done by, for example, women or people from ethnic minority backgrounds, tends to be lower status, or to be ascribed lower status, and is lower paid. Although the organisational issue within very large organisations is important, this is also quite eye-catching and can seem like a useful target, and in some ways it is. However, a really big question for the UK as a whole and for productivity is how we can tackle occupational segregation. That is a huge question, clearly, but it should be in there.

There is a conceptual problem with ethnicity pay gap reporting, but also a very practical problem. The conceptual problem applies to all pay gaps, in that it is presupposed that differences are unfair and result from discrimination or other unfair treatment, rather than from the freely expressed choices and preferences of individuals. But with ethnicity, in particular, there is a problem. It is simply not the case that the country can simply be divided into one privileged ethnicity—presumably white—and other ethnic minorities. If you look at hourly pay in 2021, for example, at the time of the last census, Asians were the highest paid salary and waged workers, at £14.29 per hour, with white people next, at £13.51, and black people next, at £12.55. This is a complex area. There is no evidence to suggest that this is happening because Asians are being fundamentally discriminated in favour of—that would seem like a nonsensical statement. The idea that we should just publish these pay gaps and then automatically see the existence of gaps as a reason to take action may well fuel discriminatory treatment, even if it is within the law.

You are talking about Asians. Are you mixing east and south-east Asians with south Asians?

If you look at the high-level classification, that is what you get. Our report, “A Portrait of Modern Britain”, goes into the sub-categories within Asian and black. We look at black African, black Caribbean, south Asian and east Asian. That information is in the report.

That would massively skew—

You absolutely see huge gaps within each of those sub-categories. I just used three here for the sake of time to show that it is not just—

It is just important that we are clear.

Yes, I agree completely. You see huge disparities between south Asian and east Asian, Pakistani and Bangladeshi. I will just say one final thing on this idea of the market determining pay. What is happening in Birmingham shows exactly what happens when you do not let the market determine pay: you have rubbish piling up on the streets because we are not being allowed to pay refuse collectors enough to get them to collect it. We saw it in the 1970s, and we have seen it in every country where the state determines wages, instead of the markets. Of course, we want protections like the minimum wage, but without the market, we end up with a failing economy and a failing society.

I have some observations on the causes of the gender pay gap and how we might look at that in analysing other pay gaps. For many organisations, the calculation of the gender pay gap was perhaps the first diversity-related indicator they focused on. What that tells us is often confused with equal pay. It tells us about a structural inequality where there is an over-representation or an under-representation of males and females in particular positions who earn particular amounts of money. The gender pay gap has almost gone away for younger men and women in a large proportion of occupations. We know why it widens: it has to do with the valuing of women’s work and with progression that does not keep pace. It is about caring responsibilities and lack of access to flexible working; pay does not progress and people do not progress when there are more demands on their time. That is what the gap tells us.

We can look to the evidence we have on how to address structural inequalities. As Peter says, barriers to progression are about promotion processes and ways of assessing and evaluating people, and there is bias within those processes. There is probably a set of interventions that we already know about that will feature in those action plans. Whether action will happen, I do not know.

There is that risk, as a number of us have touched on. It goes back to the point about what problem we are trying to solve, or what it is that we most need to focus on. Certainly, if we are going to ask businesses to collect this information, we have to give them very clear guidance. As we have all acknowledged, ethnicity is a harder thing, and disability is an even harder thing, to measure. So what guidance should we provide, and what outcome are we seeking? We all know that it is not an easy time for businesses in the current context. They worry about the additional burden of reporting or analysis and aspects of employment legislation. So we have to be very careful about how this would be positioned if we were to progress it.

First, it is important to consult businesses to understand what they do. We have evidence. There are businesses, including our own, that report on these things—I am sure BNP Paribas does too. So there are some good examples out there. What is important is to have consultation and discussion as to what has worked and what benefit it drove when they did this reporting. To your earlier question, if we have some issues, it is not necessarily about the equality of pay for doing the same job; it is about progression. So what actions can we be thinking about in terms of driving better progression?

Moving back a little to think about the complexity around measurement—especially with ethnicity pay gaps—we probably need to introduce social class as one of the key elements, because those intersections are critical in helping to understand and navigate that complexity.

I have a few questions on socioeconomic background. Louise, to what extent do you think socioeconomic diversity and inclusion is being left behind in debates about EDI?

It was very much left behind until relatively recently. That has changed over the last 10 or 15 years, particularly in large organisations; socioeconomic diversity and inclusion has come on to the agenda much more firmly. Initially a lot of the focus has been in large organisations, and particularly so-called elite occupations and professions such as law, accountancy and investment banks. Initially, the emphasis was very much on access and making sure that more people from less advanced backgrounds came into those organisations, because they were very under-represented. For some of the more progressive organisations, the focus now includes socioeconomic equity and diversity in relation to career progression. That is really important because we know that there is not just an ethnicity pay gap but a substantial class pay gap. We know from plenty of data now that people from more advantaged backgrounds often tend, although not always, to progress more quickly in their careers within those organisations, and those differences can be amplified when we think about intersections with gender and ethnicity. So, in larger organisations, we are seeing much more attention to this now, and that is a really positive step, but of course there is a lot more we could do.

Why are you sceptical about conventional social mobility initiatives? You have spoken about how they focus too much on the perceived deficits of individuals, rather than the structures that exclude people. Are there any more effective alternatives?

To some extent, we have to think about terminology. I would not necessarily put socioeconomic diversity within organisations as an agenda. It is related to social mobility, but social mobility is and should be much wider than that. One of the challenges in the UK over the last 10, 20, or 30 years is that—for a long period, up until the turn of the century—social class was not really on the agenda at all. It did return to the agenda, but in the format of looking particularly at what we call long-range upward social mobility. I am trying not to give the whole history of this idea, so I will do this as quickly as I can, but that related a lot to the work of the Social Mobility Commission from around 2010, the focus of which initially was very much on access to elite professions. It focused very heavily on helping talented, often young, individuals move upwards from—I hate these pejorative terms—the bottom of society socioeconomically towards the top. A large part of the agenda that has been called social mobility has been focused on that. Up to a point, those initiatives have had some success, although perhaps not as much as we would like. The difficulty is that that has been, to some extent, our social mobility agenda, and that sort of individualistic agenda can have absolutely no, or the most limited, impact on overall rates of social mobility. We know that in the UK rates of social mobility have not really changed, despite the fact that we have this really significant social mobility industry focused on that. My argument would be that an emphasis on social mobility, however flawed, is quite convenient at a point when structural inequalities are getting wider. It acts as a distraction because it tends to suggest that if we can argue that society is much more meritocratic, it does not matter how wide those inequalities get, because we can justify them on the basis that the most talented and amazing people get opportunities and therefore it is fair. So there are questions about fairness. To try to get to the point, it is not that I do not think we should focus on opening access to elite organisations and improving rates of career progression, but I would argue very strongly that our focus on social class and social mobility absolutely should not stop there. We need to look at structural inequalities in the labour market more generally.

In your view, would enacting the Equality Act socioeconomic duty be effective in promoting equality, diversity and the inclusion of people from working-class backgrounds?

It might be marginally effective. I will be honest and say that the socioeconomic duty—particularly in public sector work—is not my area of specialism. There is some risk that it focuses too heavily on process and not on outcome, but I would suggest that it has not really been properly enacted at all yet, so it is a little early to say. It is probably better than nothing, but possibly not sufficient to drive more substantial change.

I have a question for you, Peter: are your members working in the public sector concerned about the additional burdens associated with implementing the socioeconomic duty? What are the practical considerations for HR departments?

To address the question more broadly, I have always made the point that the public sector, of all organisations, should represent the communities it is part of and that it serves. In my experience of working across the civil service and other public sector bodies, I think that that principle is broadly understood, but the challenge is in opening up access to work and opportunities for people from more disadvantaged backgrounds to find work, whether that is through their first work experience or an internship, all the way through to employment. The DWP has looked at different channels into work, such as jobcentres, which frankly are no longer performing a purpose in helping people into work. When you talk to employers, they have not been focusing so much on these parts of our communities. We know that one of our biggest challenges nationally is the employment rate. A large number of people are economically inactive, and there may well be a higher preponderance of people who have almost given up on work because they find it so difficult to access. As a broad policy, and just trying to create organisations that can access all the skills and talent they need and to create representative workforces, along with all the arguments we have made for why these subjects are important, we have to look at routes and access to work. There are some encouraging signs; organisations are talking a lot more about essential skills—it is not just, “Give me a long list of qualifications.” It is about communication skills, aptitude for learning, collaboration and critical thinking, which is becoming more and more important. You can find those skills in people from all backgrounds. Those are the reasons why—particularly in the public sector—looking at these access routes, providing these channels and providing support to people from disadvantaged backgrounds is an important part of the process. The final point is why this has had less attention, and that is probably because it is not a protected characteristic. People will tend—to some degree, but not always—to follow what they are asked to measure. I am not necessarily advocating that this should be something that is measured, but we know that one of the challenges when people and organisations do try to focus on these things is, “What should I be measuring? How do I measure and understand these things?”

I just want to bring you back in, Ama, because I am conscious that you have not had the opportunity to speak recently. Do you share the scepticism on social mobility initiatives? Do you think they are an effective tool, or are there alternatives you would recommend?

Social mobility is challenging; speaking in terms of a global organisation, class is a very UK-dominated thing. If you speak to Europeans or Americans, they know that class exists, but it is very difficult to categorise. The reason I say that is that we measure social mobility. In order to decide how to gather that data, we asked the questions that have been advised by the Social Mobility Commission. One of the first questions was, “Did you have free school lunches?” As we are a global organisation, some people responded, “I grew up in Paris. Everybody had free school lunches. How does that indicate that I was in poverty?” Another question we kept coming up with was, “Did you have free education, or was your education paid for?” I grew up in Belgium, where all the good schools are state schools; if you went to a private school, it indicated that there was something quite unique about you. So the first thing I want to highlight is that measuring this is really challenging, especially within a global organisation. The second thing I would highlight is that social mobility is very intersectional. In terms of this idea that you can just focus on people from a lower class and how you bring them into an organisation, it is quite difficult to assess and measure these things. The third thing I would like to mention is that people tend to grow out of their social class. What I mean by that is that we have a few senior leaders from working-class backgrounds, but they have lost their accents and the way they would have spoken. You would only know that they were from a lower-class background if they told you. Another thing to layer on to that is that not everyone is proud of being socially mobile. When you add up all these factors, it becomes really difficult to measure social mobility. The question you then have to ask is, what are you trying to achieve? Do you want to bring people from lower classes into an organisation so that you make it more equitable, or are you trying to bring people from lower classes into an organisation so that they can improve their class? There are a lot of things that people do not think about when they are discussing what they are trying to accomplish through this conversation about social mobility. It is necessary, but it is complicated. Those are the things that I come across and think through when we are considering social mobility.

I have a couple of very quick points on that. I completely agree that it is complex, but I do think we know what we are trying to measure. If we look at organisations in the City of London and the UK—although this does exist worldwide—we know that, even now, they are heavily dominated by senior leaders who are predominantly from upper middle-class backgrounds, predominantly male and predominantly white. That has not changed as much as the City of London would like us to believe for a very long time. In terms of measurement, we can measure who is occupying those positions and the extent to which that changes over time, and that is the target of organisations such as Progress Together. As a sociologist I have to mention that there is a fundamental tension—in fact, there are many tensions. First, we have noted that class is slippery, subjective and difficult to measure; we change class throughout our lifetime. It is undoubtedly complex, but what is also complex is that much of the social mobility agenda has been focused on very large organisations such as banks, accountancies and law firms. The difficulty, tension or paradox for some people like me is that these are the very places where social class is made. If we think of social class as hierarchies in terms of income, wealth and money, our society is extraordinarily hierarchical, and that hierarchy is partly based on the income inequalities that are made in the City of London. So what we are trying to do with this agenda is shuffle the pack a little in terms of who gets those opportunities, but that fundamental class structure, which is defined by material inequalities, stays in place. We can argue that, in shuffling that pack to a little extent, we actually legitimate that underlying class structure; we make it more legitimate, safer and more structural, as opposed to changing it. It is almost an elephant in the room in terms of that agenda, but it is an important point to raise.

Alex is going to move on to something that nobody can argue we cannot measure: the implications of an ageing workforce.

This has been fascinating; thank you all. Zoe, do you think age diversity and inclusion features strongly enough in our EDI strategies, given high youth unemployment and the UK’s ageing workforce and the opportunities and challenges those bring?

I think it is relatively under-attended. The risk is that we treat it as a demographic fact rather than a source of inequality in how opportunity and contribution are distributed over the life course. It shows up as a source of inequality in key decisions about individuals’ access to jobs and training, about progression and about how capability is judged. We see this at both ends of the workforce, with older workers facing assumptions about adaptability and the adoption of technology, and being overlooked for development opportunities at that later life and career stage, while younger workers have limited influence and access to progression. In practice, day-to-day inside a workplace, intergenerational teams are a reality. There are differences and tensions around how experience is valued, what communication styles and preferences there are, how feedback is given and who gets heard. Where that is not managed well, it can translate into exclusion. From a Government perspective, there are probably two obvious gaps: we have limited data on progression by age and training by age. We actually have less clarity on what good looks like in terms of age distribution within a workplace or a workforce than we do with other characteristics. We talk about representation, but how does that work with age? So, yes, it is an equality issue. There are definitely some new challenges around workforce effectiveness that speak to the productivity point of a multi-generational workforce, and that will only become more pressing.

Ama, I wonder if this is an area that you address in your business and what the key features of an effective approach to maximising the benefits of a multi-generational workforce might be.

I would have to be direct and say that we look at age in terms of the younger generation. We spend a lot of time on our graduate programmes, with our analysts and with our apprentices. To be blunt, those measurements are easier because you are dealing with universities, schools and institutions that are quite easy to access, and these people are quite easy to upskill. Where we really have challenges is with the older generation, in terms of upskilling them and getting them used to the changing world of work; I am talking about AI, remote working and all the structural changes that have happened in the workforce. We have to spend more time and energy in terms of the older workforce. I do not have a long list of things that we have done for the older workforce, although I have a long list for the graduates. Most of you will be familiar with that: training, opportunities in terms of talent programmes and talent development, recruiting from a wider range of universities and whatnot. I cannot necessarily say the same for the older workforce. The obvious things that come to mind are reskilling, talent progression and promotion, and those become more complicated when people have different skills and abilities. I do not have a list of things that we are doing or achieving that are making or driving change. We are very much learning; as it becomes more obvious that the intergenerational workforce is the way forward, organisations are going to be challenged as to how they manage their older generational workers, who are still skilled and still keen, but who may have abilities or strengths that are being challenged or being developed by other ways of working or other instruments of working.

Can I quickly come in? We have talked about the older workforce, but do we all agree on what counts as the older workforce? What age group would you classify as belonging to an older workforce?

It depends a bit on the context. There are definitely sectors—banking being one probably, and tech being another—where people in their 50s would be counted as older workers, but there are other environments where people in their 60s would be counted more as older workers. The shift in demographics is definitely a long-standing problem, but there is also the issue of older workers needing access to work. That is good for their wellbeing, but also in terms of the economic impact, because many older workers have not saved up enough for their longer-term retirement. Businesses can look at a number of things, and Ama has touched on a couple. Flexible working is an obvious one. There is also access to good wellbeing environments, which actually benefits all. Again, we are coming back to inclusion ideas, and these are not necessarily directed just at older people, but they can help them. There is definitely an emerging problem of the need to upskill and reskill. There is the assumption—this is an inbuilt bias that I guess we all have—that you cannot teach an old dog new tricks. In fact, many older workers prove themselves to be highly adaptable, and when they are given the opportunity, they can also prove to be very loyal. Going back to the point about representing a customer group, we have seen some very good examples of organisations that deliberately recruit older workers because they know that a lot of their customers are from that demographic. So there is still more that we need to be doing in this space.

Just to share a definition of an older worker with you, the Centre for Ageing Better, one of the What Works Centres, has a focus on work and employment and defines an older worker as 50-plus.

That is savage. Does everybody else agree with that?

It is sector-dependent. When I worked in professional services, I was told I was retiring at 52.

Having worked in professional services, I know that the retirement model in that environment is from 50.

Given that the retirement age is 67, if the definition is 50-plus and if you take the first 17 years as you being a younger worker, there is not much left in the middle, is there?

One of the biggest changes for older workers has been around older women workers and menopause awareness. We know that one in 10 women leave the workforce when they reach menopausal age. We are talking about an older worker being 50-plus, and some may say that such changes are a regulatory burden, but how much are they going to enable us to ensure that older women workers can stay in the workplace?

There absolutely should be a stronger focus on menopause and health through the life course, as well as an understanding of the different needs of workers at different life stages in terms of commitments outside work. All of that intersects with age; it is an intersectional issue that really has not had as much focus. It is benefiting from some now, but there is much more to do.

I would add that there is a need for employers to be more open and more able to recognise the idea of non-linear careers. Often the career model has been based on the traditional male model of permanent employment and career progression upwards with no significant interruptions; that model has been the norm against which difference is measured. However, more and more people are having longer careers, interrupted careers, and are changing jobs. That non-linearity—some people call it the labyrinth, as opposed to the glass ceiling—is a really important trend that employers need to be able to get their heads around.

They talk about squiggly careers now—I have heard that expression.

I would endorse that, and some of what Peter said about the variety of skills. Just in terms of whether age is neglected, as part of our report we carried out an EDI review of FTSE 100 websites. While I do not have an actual number on each of the characteristics, certainly anecdotally age was dramatically less covered than sex, race or sexual orientation. It may have been covered in recruitment practices and reskilling, but there seemed to be less focus. One thing I would add on labyrinth careers, which also applies to the previous discussion on social mobility and socioeconomic diversity that Nadia mentioned, is that some employers in the public and private sector automatically ask for a degree when it is not specifically required. First, we know that there is a huge gap between the top and bottom quintiles: the top has almost double the higher education participation rate than the bottom. We also know that this is highly dependent on age: workers in their 50s are much less likely to be graduates, but they may have a huge amount of career experience. Removing that degree requirement unless it is essential—such as for a nurse or an engineer—is a simple thing that employers can look at if they want to broaden their diversity in both those areas.

Sometimes degrees can be required even when somebody has many years of experience, and that is mad.

That tallies with our work on older people in our previous sessions, particularly around discrimination at work. We found that people were discriminated against for their age, but there were very few employment tribunals or people who took those issues forward. I am going to move on now to Christine and disability inclusion.

Can I just say how informative this has been? I have learned that I am an older worker who has had a labyrinth or squiggly career.

And you are valued and experienced.

Moving on to disability, how can EDI practice best support disabled people’s recruitment, retention and progression at work?

I am going to start with what I have started with quite a bit: disability is very challenging. Visible disability can obviously be seen, so it is easy to support. However, invisible disability is very difficult to assess, and you find that there are biases, especially in the workplace, associated with disabilities. Coming from a finance background, I am going to talk about neurodiversity because that is the biggest category of disability that I see. Actually, that is seen as a strength, but we have really low levels of reporting among our employees in terms of invisible disability. To answer your question directly, with regard to EDI and disabilities, employers have to raise awareness and create an environment and a space where people with disabilities feel that they are supported and that their disability will not hinder their progression and their development. I would say that is the best way to approach it. The classical measuring targets and setting up KPIs in disability is quite challenging; there is a real aversion in the private sector to reporting and sharing that information, and it is very hard to measure and drive progress with something that you cannot see.

Zoe, was there anything you wanted to add?

I am not a specialist in this area, but I am on flexible and hybrid working, which is potentially a huge enabler to increased employment of disabled people, particularly those with disabilities that may require an adjustment to the time, timing or place of work. I would like to speak to some changes in that space that may assist. We have seen a real expansion in the flexible working that is formally available and in the adoption of hybrid working, particularly in formerly office-based jobs and, frankly, at the higher-paid end of things. However, the evidence, including that from my own ESRC-funded research work after lockdown with the University of Southampton and the Institute of Employment Studies, is that access to flexible working is not yet experienced as a consistent entitlement. The core issue is actually quite basic: most organisations have not systematically analysed what flexibility is possible in different roles, and they do not communicate it clearly. Where that analysis is absent, access to flexibility is decided case by case and usually by line managers, which creates uncertainty for candidates. Particularly for disabled people looking to avail themselves of the right to flexibility from day one, that uncertainty creates a real barrier to entry. What matters in practice is clarity in a very small number of dimensions: hours, timing, location of work and how stable or variable that pattern of employment is over time. Is it different every week, or is it the same? If that could be analysed and published in relation to every job role, it would sort a lot of things out. If organisations were expected to set that information out clearly in job design and job adverts, it would remove a significant amount of uncertainty which inhibits people—particularly people with a range of disabilities—from accessing employment.

I was going to come on to flexibility. There was a cultural shift towards flexible working during and after covid, which a lot of us presumed would benefit disabled people in the labour market. From what you are saying, that has not emerged in the way that we expected. Peter, why has that not yet had a substantial impact on the employment rate of disabled people in the workforce?

It is probably helpful to say what we mean by flexible working. There is place, location, time and so forth; it is not just about hybrid working and working from home. It is also about part-time working, job sharing and a whole variety of things. What became very clear through covid and subsequently is that this is what a lot of people are looking for in their workplace, so I agree with Zoe’s point. At the CIPD we have long advocated that people should see flexible working as part of an employee value proposition. We know that, for a whole variety of reasons, many people are interested in the possibilities of flexible working, in the different forms I have described. So that may be part of it. Organisations did not of themselves immediately think, “Great, that is exactly what we need to do for people with disabilities.” It has been acknowledged that there is still bias in terms of understanding disability; we have all talked about that, and particularly the invisible side of disability. Businesses worry about the cost of recruitment and potential reasonable adjustments. But it comes back, as we have stressed throughout, to what the value is to business: why would this help them? We know that across the working population, 20% to 25%—pick a number—have some form of recognised disability, including neurodiverse disabilities. So it has to begin with helping people to understand what the benefit is of recruiting disabled people, and it comes back to the wider inclusion idea. I agree that flexible working is one of the obvious ways in which we can support people with disabilities into work, and we should be looking at that directly. But as you know, in the Employment Rights Act there is more of an onus on employers to say, “You’ve got to talk about this.” If somebody asks to be able to work more flexibly, there need to be very clear reasons as to why you would prevent that. The final point I would make in all this, and it remains a resonant point, is that we have to train our managers better. If managers do not understand these things, they will not know how to deal with people with different forms of ability or the diversity we have discussed, or how to deal fairly with people who work in different ways. We know, for example, that there is an inherent bias against part-time workers; in the public media we sometimes hear that people who work part time are somehow less committed; you get this argument going on. Fundamentally, we need to drive not just this aspect of inclusion, but all aspects of inclusion, and to do a better job of training managers so that they understand these things better as well.

There have been a lot of back-to-work mandates from offices since covid and lockdown. What impact might that be having on disabled workers? Is it having a greater impact on them than on the rest of the workforce?

It could be; it is an interesting one. Working from home and hybrid working in particular are very context-specific, and every organisation has to figure out what works best for it. They should be consulting their workforces and recognise that there is not necessarily only one model for the whole organisation. If I am honest, these mandates have often caused a bit of reaction. The workforce is saying, “Why are you telling me I must do this when I have been able to show that I am productive, and I can also tell you that it is good for my wellbeing and my loyalty?” These are the things we should be looking at as part of why we create more flexible ways of working. We are still, arguably, learning from the covid crisis, which was a big accelerant of these ideas. But we would always advocate for employers to consult with workforce and figure out what works best for all. If you are going to mandate, you had better be clear as to why: what is your evidence that it improves productivity or any of these other things? Again, you have heard a resonant message that building the evidence case for the practices that we put in place, whether supporting disability or anything else, remains an imperative.

Could we circle back to something I was going to ask before? To what extent can employers’ accreditation schemes such as Disability Confident help change employers’ attitudes towards recruiting and encourage them to think more about flexible working?

The challenge is the sustainability of some of these things. When we first saw Disability Confident, many organisations argued that it shifted their behaviours. But over time the worry is that it is a virtue signal—I have the badge on the door—but where is the evidence of practice that has followed it? What are my reporting obligations towards it? That is part of the challenge of these things. I keep coming back to the idea that we should articulate the value to enterprises and that they should start to build evidence of the value it is creating into their own business performance and metrics. We are seeing more interest from other stakeholders, and not just regulators, and I am sure that Ama would report that too. Business investors are increasingly interested in what you are doing to create a sustainable workforce that has the skills and capabilities, that has a good culture and that is driving outcomes like productivity. We need to see more of that. It goes all the way back to the question that Kevin asked right at the very beginning: what is the role of Government? Sometimes it is to do a little more compliance, but compliance will never solve all these things. It is also about guidance and bringing people together to say, “Let’s understand the benefits of this and work together to create the benefits case.”

My observation is that schemes and kitemarks around Disability Confident and Access to Work are helpful to a degree, but actually it is the detail of the job design that is going to support an individual and their particular needs. I go back to the point about how the job is designed, how interdependent it is task-wise in terms of collaboration with other people, and how that is enabled through technology. The signalling may communicate that individuals with disability—or acquired disability, in the case of an employee over their time with an employer—can disclose and that the employer will support them. But I would say that, in reality, the practice, sensitivity, adaptations and adjustments have not really followed at the same pace and confidence for people.

I am going to make a potentially slightly controversial point: pressure from Government, kitemarks, charters, the business case all contribute to change and to employers making changes. The biggest challenge is to move this from important to urgent because it is the more urgent things that get addressed. I would always stand by the argument that—although it is difficult to say this, and it might not be popular with some organisations—reputational pressure is massive and actually far more important in focusing minds on driving change than other more traditional metrics or drivers that we might talk about. It is a driver, but it is also part of the problem. It is a driver because it drives change, but it is part of the problem because once the reputational issue is solved, so too is the problem; it has gone away at that point, and that does not always lead to fundamental, underlying, systemic or structural change. We also know that once reputational pressures fall away, then so too do DEI measures. We saw, for example, with the Black Lives Matter movement that there are sudden activities after a particular event because it is reputationally important, but then the reputational driver falls away and so too does that activity. I would just put that in there, because I think we can use reputation. It is slightly problematic, but it is also a driver and a lever for change at the same time.

I would add that reputation drives a lot of things that organisations do, but if you look below that, it is really what your clients and your investors want you to do. Yes, there is a challenge in that once you have saved your reputation or your reputation is fine, there is not really an onus to do the work, but where the pressure does come through is through society. I would argue that what the younger generation does very well is that they are very invested in what an organisation looks like, what its culture looks like and how it treats its people, and that helps drive change within organisations. I also think that clients and investors matter. There is a lot of talk about greenwashing and DEI washing. All these “washing” labels are about investors, clients and customers holding institutions to account. That helps to drive change in a more significant, sustainable way than just an internal focus. That is not to say that Government does not have a role to play, but these things all interlock and work with each other, especially in the space of disability, which can be an invisible trait, so you need other pressures because it is easy to ignore. If you are a woman, that is fairly obvious, and if you are from a particular race, you can tell that to a certain degree, but with disability, the challenge is that it can be invisible.

Iain, I have a quick question for you. I know that when we had the discussion about welfare reform, Policy Exchange argued that, for 16 to 30-year-olds, PIP should be conditional on looking for work. Is there not merit in saying that workplaces should be more disabled access-friendly, if we are requiring disabled people to find work and be in work to receive benefits?

I am not a specialist on this; my colleague Jean-André Prager wrote that report. But this is related to a point I was going to make, and I will make sure I address both together. There is a danger when we look at disability in the workplace that we focus on being inclusive to those who are already in work. Obviously, that is a good thing to do, but we ignore those who are out of work. We know that disabled people are much more likely not to be in work, and sometimes not seek work, according to the conditions placed on them. Particularly for young people, some of the largest growth in UC claimants without “seeking work” capability comes from disability, particularly around mental health and neurodivergence. We have to be really careful of second-order effects. We want to make businesses more accommodating, more friendly and more welcoming to disabled people and to help them into work, but we need to make sure we do not put on requirements that make them less willing to take on disabled people and to take a chance. We know that this is something where you get second-order effects and where employers, particularly small employers, may choose not to comply, whatever the law says. Again, I do not want to go into too much detail as I am not the expert, but the thrust of the previous report that we wrote, and what I would say here, is that we essentially need a twin-track thing of higher conditionality for 16 to 30-year-olds applying for PIP in terms of seeking work, although we also said voluntary activities or training. That needs to be combined with pressure on employers to obey the law and potentially to look at how they can go further in getting disabled people back into the workplace. The impact of scarring, if you are under 25 and out of the workplace for 12 or 18 months, is huge and can have lifelong consequences. This is an area where we need to push from both sides; I would be worried about something that only pushed from one side or the other.

I want to briefly address learning disability because it is very much at the bottom of everyone’s pile of things to look at when it comes to diversity and inclusion. Only 6% of people with a learning disability are in any kind of paid employment, and this has been the case for a very long time; it is an intractable problem. Yet, there are 40,000 people with Down’s syndrome in this country, and plenty of them are able to work—maybe not every career is suitable, but that goes for most of us. Whose responsibility is it? How do we start to address this problem? We know that being in work adds a sense of purpose; you have talked about the damage that can be done if young people are out of work for a long period of time. How can we start to address this? Is it individuals’, business’s or Government’s responsibility, or a combination of all of those?

I would like to say something that is very related to that. It is a worry that we can sometimes look at those who are least in need of help. Take the case of a middle-class professional who is diagnosed with ADHD well into their career; we like to think of them as differently able and neurodivergent. Yes, there is some value to those perspectives, but it can end up being surface-level inclusion, with us seeing these things in a way as superpowers. But when we look at those with Down’s syndrome, who are very differently abled, there are things they can do, and many things they cannot do; they can just be forgotten about, and that’s that. You mentioned the 6%, so the numbers we are talking about are enormous. This is an area where I worry. When I was much younger, I worked in a supermarket that employed a guy with Down’s Syndrome who collected the trolleys in the car park. It was great for him; he earned a wage but it probably added minimal value as part of their social contribution. Would I want them to be dinged for a disability pay gap because they employed him, and then to get rid of him to improve their disability pay gap? I am not suggesting they would have done that, but these are the sort of second-order pressures I am talking about. As I said, we have done work on this. These people are vulnerable in society, and it benefits everyone if they can make some contribution, but Government, civil society and employers have to work together on this one. It is an area where Government can try to work much more collaboratively, not through tick-box exercises or heavy burdens, but by having a real expectation on employers to join in with these sorts of incentives.

I would just add that collecting trolleys at a supermarket is not minimal productivity; somebody has to do it, and the company would need that, full stop. Zoe, did you want to add anything?

What is needed is a combined approach—importantly, led by Government. There may be learnings in criminal justice models, in terms of opportunities for ex-offenders. There are particular very vulnerable groups in terms of social exclusion and access to life opportunities, for which meaningful employment—that might be trolleys, or it might be something else—is valuable. So this absolutely should be in focus.

We have talked about the business case, about CSR, ESG and all those agendas, and about a version of stakeholder capitalism that is not purely about shareholders. There needs to be room to underline that employers, and large employers in particular, have a duty to society that may be quite separate from a business case, because society is what enables them to make their profits. The more we can make that argument, the better, because that is where joint activity can work more effectively. I do not want to undermine the need to make money or the business case, but if we constantly hold the conversation only in those terms, it can be extremely limiting.

That was a really important question to finish on; I know it is one that Alex holds dear, as does Kevin, who would have been sad that he missed that contribution and discussion. Thank you again for coming back, and thank you to Ama for joining us. We are very grateful. We have managed to get through without any votes, which is a good thing—don’t worry, they are stored up for later. That brings this session to a close.