Women and Equalities Committee
Oral evidence: Race Disparity Audit, HC 562
Wednesday 7 February 2018
Ordered by the House of Commons to be published on 7 February 2018.
Members present: Mrs Maria Miller (Chair); Angela Crawley; Philip Davies, Eddie Hughes; Jess Phillips; Tulip Siddiq
Questions 92 - 152
Witnesses
I: Melanie Field, Executive Director, Corporate Strategy and Policy Wales, Equality and Human Rights Commission; Emma Rourke, Director of Public Policy Analysis, Office for National Statistics; Iain Bell, Deputy National Statistician, Office for National Statistics.
II: Rt. Hon David Lidington CBE MP, Minister for the Cabinet Office; Marcus Bell, Director, Race Disparity Audit, Cabinet Office.
Written evidence from witnesses:
RDA0011 Equality and Human Rights Commission
RDA0027 Office for National Statistics
Examination of witnesses
Witnesses: Melanie Field, Emma Rourke and Iain Bell.
Q92 Chair: Thank you so much for being with us today. Can I say good morning to you and to anybody who is watching online or in the public gallery? This is the third and last oral evidence session in our inquiry into the Government’s race disparity audit, which has presented to Government data about disparities between different ethnic groups in different areas of life and public services.
We have two panels of witnesses today. In a moment, after we have had this session, we will be hearing from David Lidington and the Cabinet Office. Our first panel here today gives us an opportunity to look more at the work of the Office for National Statistics and the Equality and Human Rights Commission in this area, and at issues that should be addressed as a result of the race disparity audit.
Before I start, could I just ask you to say your name and the organisation that you represent?
Melanie Field: I am Melanie Field from the Equality and Human Rights Commission.
Iain Bell: I am Iain Bell, deputy national statistician for population and public policy at the Office for National Statistics.
Emma Rourke: I am Emma Rourke from the Office for National Statistics.
Chair: Brilliant. Thank you all for being here today.
Q93 Tulip Siddiq: Good morning. Thanks for coming in. My question is to Melanie to start off. The EHRC’s submission to our inquiry seems to suggest that you think the work that the organisation is doing is better at uncovering inequalities than the race disparity audit. In your opinion, do you think the audit has been a waste of time and resource?
Melanie Field: No, I do not think the audit has been a waste of time. We do our own work and have a statutory duty to report to Parliament on progress towards equality and human rights in Britain. We do that through our “Is Britain Fairer?” report. That is a very structured approach, for which we use a measurement framework that looks at all areas of life and uses a variety of qualitative and quantitative data sources.
The race disparity audit is a different exercise and fulfils a different purpose. It is able to do things that our approach is not able to do, for example in terms of the amount of regional analysis, but on the other hand it has gaps where we do publish data. It fulfils a different purpose. It has also been extremely beneficial that the Government have shown such an interest in this issue, and it has certainly raised the profile of race disparities in society, so we welcome that.
Q94 Tulip Siddiq: The EHRC said that the data lacks completeness, robustness, consistency and comparability. Is this something that can be remedied and do you have some suggestions about making the audit comprehensive?
Melanie Field: Yes, things can always be improved, and that is probably equally true for our own work. We always look to improve. We talk to the race disparity unit on an ongoing basis about where we think additional data would be useful. I know that it is planning to add more data to the website and we will be seeking to guide it in that. For example, on higher education, we think that there is more data that is available and could be used.
The main issue for us is around putting the data in context. It would make the data more useful if there was more contextual information that enabled people to understand what it means, and that drew links between different datasets, for example looking at the relationship between access to and attainment in higher education, and then linking that through to employment.
Q95 Tulip Siddiq: Are there lessons that we should be learning from the audit in terms of how Government collate data and how we can make it more relevant, robust and useful?
Melanie Field: Certainly, and I am sure that ONS will have views on that as well. There are differences in the ways that organisations disaggregate ethnicity data, which make it difficult to make comparisons. Change in the way people do things over time makes it difficult to track progress. There are lessons to learn in terms of completeness and consistency about how we look at these issues.
Q96 Tulip Siddiq: I am just going to push you a bit further on the usefulness of the data. You have already explained about having some kind of context, which makes sense. Is there any other way the data could have been presented that would have made the data more useful to us?
Melanie Field: The Government took a view that they were going to publish the raw data, in effect, without putting any analysis or context around it. I can see that there are risks for Government if it is not just presented in its raw form, because the way that the information is presented may be seen as being political spin, but for people to use the data they need that analysis and context. We try to provide that with our report. Our report is on a different basis, but there is room for the Government to supplement what is there with something that would make it easier for people to use.
Q97 Tulip Siddiq: Iain, I do not know if you wanted to come in on this point.
Iain Bell: Sorry, on the point around the usefulness and comparability?
Tulip Siddiq: Yes.
Iain Bell: The way the system hopefully works is that we publish in ONS what is called “harmonisation guidance” and the departments, when they are collecting the administrative data for how they run benefits or education systems, and in the surveys they conduct, are then advised to harmonise against those classifications. What became apparent through the race disparity audit is something that was known, but it shone a light into this area: that many different organisations have yet to align to the latest ONS classifications for the 2011 census, and of course we are now coming up to the 2021 census.
As part of the preparation for the 2021 census, we are currently working through what the ethnicity classifications should look like, but we know that there is a real job to do to work with organisations across Government in order to get that consistency of approach right the way across all the data collectors, and make sure that the harmonisation is there. Some NHS bodies are collecting 2001 definitions still. I know the Department for Education, my old department, was on a hybrid between 2001 and 2011. If we can get everybody up to the same level and work with them as part of the progress towards 2021 census, it will lead to an increase in comparability across the board. I view it as being a task I now have to take on, as part of the next stage of the census development.
Q98 Chair: Can I just ask Melanie one supplementary before we move on? The EHRC said that it was worried about the robustness and consistency of the data. Are there any particular examples or areas where the organisation was concerned about the robustness of the data? That is worrying. I can understand completeness and comparability but, if you are saying that some of the data in there was not robust, it would be useful to have a particular example.
Melanie Field: I am going to have to write to you about that.
Chair: Is that okay? Would you be able to?
Melanie Field: Yes, of course.
Chair: Thank you very much.
Tulip Siddiq: Maybe when you write you could put in something about how you would have done it differently to make the data more robust, because that is what we are trying to get at. If an audit like this is done, how do we trust the data? To say that it lacks completeness or robustness makes us question the integrity and correctness of the data. If you could write to us, that would be helpful.
Melanie Field: Certainly.
Q99 Eddie Hughes: Melanie, where should responsibility sit in the Government for ensuring actions on the outcome of the audit?
Melanie Field: It is really important that there is a central point that drives a strategic approach across Government to addressing the issues that occur across all areas of life in relation to race equality. I understand that, following the recent Cabinet reshuffle, responsibility for the race disparity unit now sits with the Cabinet Office, as it has a policy co‑ordination role. Previously it was a split responsibility. That is really welcome, and I hope that it is a strong push from the centre to take co‑ordinated action across all these issues. In fact we would like to see that happen for all of the equality‑protected characteristics that have different responsibilities across Government. It does not matter which department it sits in, as long as it sits in a place that has an overview and is able to influence action across Government.
Q100 Eddie Hughes: What do you think about the action that has been taken so far by the Government, devolving it out to the relevant departments?
Melanie Field: There is a push and pull thing going on here, and that is probably right. You need that helicopter view to say what the big, priority issues are that we really need to focus on getting some traction on. Then departments that are responsible for particular issues are best placed to come up with policy responses, but we have not yet seen what those policy responses are going to be.
Q101 Eddie Hughes: What is the role of the commission in all of this?
Melanie Field: The role of the commission is to form our own views about what we think the priorities are, based on our analysis and understanding of the data, and then to encourage and press Government to turn the information into real action.
Q102 Jess Phillips: I am afraid this is to you again, Melanie. I feel like we are picking on you now. The UK has been collecting ethnicity data to identify inequalities for a good many years and, while there have been some successes in reducing inequalities in a few areas, the audit shows that some measures remain stubbornly entrenched. Is it reasonable for us and the EHRC to expect that this initiative and the race disparity audit will make a significant inroad in tackling the more entrenched problems?
Melanie Field: To answer that, you need to think about what conditions you need to make progress on an issue. Transparency, leadership and commitment are part of that and are the beginning of that journey. What then needs to happen is for people to take ownership of the issues and act on them. We have known about a lot of these issues for a very long time and they persist. Progress has not been as rapid as we would like to see.
Q103 Jess Phillips: Is knowing it a little bit more going to change anything?
Melanie Field: There is a risk. I completely support that action should be based on evidence, but there is a risk that a search for more and better evidence can delay action. If we know that certain ethnic groups are underachieving at school, it is useful to know all the factors that contribute to that, but it does not stop you trying to do something in the classroom.
Q104 Chair: Melanie, you said that the EHRC is going to form its own response to the audit. What form will that take and when will we see it?
Melanie Field: We published our roadmap to race equality shortly after the audit was published, which identified a number of areas where we felt that urgent action was needed. We worked with some race equality organisations to produce that report.
Q105 Chair: Is there anything else that you are planning to do as a result of that?
Melanie Field: We are continuing to keep an eye on what is happening. We are in touch with the race disparity unit, and we are continuing to ask it how the policy response is progressing. Another thing that we are doing at the moment is research into the public sector equality duty. That duty obviously followed on from the earlier race equality duty. The policy idea behind it was that it should harness the power of the public sector in addressing persistent equality issues. It is clear from the figures that that has not been entirely successful, so we are looking at how the duty has been working and considering whether we can make some recommendations about how it could be improved to make it more effective.
Q106 Tulip Siddiq: Did any of the data from the audit surprise you in terms of inequalities or was it all old ground that you expected and had seen before?
Melanie Field: I do not think that anything came as a massive surprise to us, no.
Q107 Angela Crawley: My questions are for the ONS. Your new audit is going to be looking at the advantages and limitations of ONS administrative and survey data on ethnicity. Does this mean that you think the race disparity audit has fatal flaws?
Iain Bell: The work programme, and I will hand to Emma to talk about it in more detail, is broader than simply looking at ethnicity. We are looking across all the protected characteristics. The aim of the audit is to provide a coherent assessment of the data that is available, not just on ethnicity, but also across all the other characteristics, on gender identity, sexual orientation, disability, et cetera. Emma, do you want to update us more on where that has got to?
Emma Rourke: Sure. The inequalities data audit that we started just with Government bodies, inviting them to look through our baseline of all data that we had available to us and add to it, has now concluded.
We have had 46 responses, which we feel is particularly reassuring because it means that for the majority of our Government stakeholders we already have sight of the data that they had, and it was being used either by us or through the race disparity audit. Then for the 46 they had something additional to offer to us. We will have a look through the quality and added value that that data can give us.
Q108 Angela Crawley: How satisfied are you that the data standards in the Government’s audit are high enough to draw policy conclusions?
Iain Bell: Do you mean on the race disparity audit?
Angela Crawley: Yes.
Iain Bell: It is worth pointing out that much of the data in the audit is already published in individual departments, and it is published often with the badges of national statistics or official statistics, which are designed to provide reassurance to users of the statistics that these are done in line with the UK Statistics Authority code of practice and are of high quality.
The audit has tied it together and provided the overview across the board, because many users who are interested in coming at this from an angle of race disparity may have had to go over several different websites or parts of websites to find this information. By making it more easily accessible, it is there. ONS also played a role, in that we seconded out some staff to work in the audit, to make sure that the quality and accuracy were there within the audit on that side.
Q109 Angela Crawley: We have covered this briefly but, based on what the audit revealed about how the public sector collects ethnicity data, could there be improvements to how the public sector collects that data?
Iain Bell: Yes, there undoubtedly is room for us to get more consistency and comparability across the board. If everybody is using the same or a similar definition, it aids the interpretation and it is easier to follow through some of the disparities we have talked about already. If the education system is not on the same definition as the data used elsewhere, say in the labour market, it becomes harder for users to quickly come to conclusions.
Q110 Angela Crawley: The improvements you would recommend there are consistently, comparability and similar definitions.
Iain Bell: Yes, alignment with the harmonised standards that we publish.
Q111 Chair: Could I just go back to the specifics of administrative data? We are not statisticians, but we understand that there are real concerns about the way that data is gathered and then used, and that the limitations of that data need to be carefully considered. I am thinking particularly about Gypsy, Roma and Traveller communities, who can feel that their invisibility in the system is compounded if that sort of data is used. Would you agree?
Iain Bell: We have different sources. Fundamentally, we have census, surveys and administrative data, all of which have different strengths and limitations. The issue you highlight with Gypsy, Roma and Travellers, as a specific example, is that, for each of these—for the census, surveys and administrative data—there are different reasons for the interaction. The census is one‑off, every 10 years and limited. There must be trust in the system and that the data is used impartially and independently. We would like to think that we run a census in that way, which enables that.
Surveys are usually run for households on a voluntary basis, and for those communities there are two issues that we look at. One is how we make sure that, within a survey of households, we are even getting to that population. Secondly, because it is voluntary and run by Government, there are issues around whether they would naturally respond to this data collection.
On administrative data, when the Gypsy, Roma and Traveller community interact with the service, the data will be collected naturally as a by‑product of it. For the vast majority of administrative systems, it is recorded based on what the person wants to express their identity as being, and relies on the person being willing to say that they want their identity to be recorded in this ethnic group.
All of them rely on the trust between how the data is collected and the body that is doing it. Particularly for the group you highlighted, it would be difficult to conclude that any one system is better than any other. Each has different strengths and limitations, and in using the different sources of data it is important that we make it clear where the strengths and limitations are, so that people know and understand the issues that can arise. It is not fair to say that administrative data is particularly poor. It is compared to others.
Q112 Chair: Given that we are dealing here with groups of people where trust may not be strong, what are you doing to ensure that new methodologies are adopted to try to overcome that? We are otherwise living with a fatal flaw in the data, are we not?
Iain Bell: As I mentioned earlier, we are preparing for the 2021 census. One of the core aspects for that is community engagement. We are already in large‑scale stakeholder engagement about the user needs for ethnicity data and the acceptability for what we include in the census.
In ONS, we have often set up this community engagement that helps all these groups have trust and fill in the census, but then that infrastructure tends to melt away a bit into the background after the census has happened. We are currently trying to design this in such a way that we do not set up a community engagement programme for giving statistical information that builds to a census and then disappears again, but one that is on an ongoing basis, so we can build that trust to provide the information, which allows us to have accurate data. It is work that we are building to at the moment, and we are determined to sustain it rather than leave it as a one‑off event.
Q113 Chair: It feels to me like you might almost have mini censuses going on between censuses, because the one big problem of a census is the gap, is it not?
Iain Bell: Yes. We have a programme of research at the minute, which is called the administrative data census. That is looking at the information that is held across Government at the moment and whether we could use that to provide more timely updates on the population on an ongoing basis in between census years. Yes, we could be looking at mini censuses, not in the sense of actually conducting a census, but mini censuses of the population using data that exists already.
Q114 Chair: Moving slightly further back to the race disparity audit—but it was important to know what other datasets might be available—much of the evidence that we have received has called for far more granularity in the data that is included in the audit, and the ability to control for variables so that it can be determined whether race is a factor in causing a particular disparity. Does the audit provide enough detail for this sort of robust analysis of the causes of disparities?
Iain Bell: There are certainly examples within the audit where it gave more information than just the ethnicity breakdown. The one example that immediately springs to mind is education, so information on free school meals was provided as an indicator of income levels, to try to get into deprivation. Where there was some data available, it provided the information available.
Fundamentally, turning to the work of Emma’s team, the importance of the work is the ability to look at all the characteristics and not just race on this side. It is our view that we need to do a bigger job, building on the work we have just started with the audit that we mentioned earlier, on all the inequalities. Emma, do you want to add a bit about the work that we are doing there?
Emma Rourke: I would just emphasise the fact that we see the race disparity audit as it currently stands as the beginning. It is providing a common evidence base from which we can build a variety of different conversations. While it does not necessarily all come as a surprise to some of us who are familiar with the content, there are a number of third parties who will not have had access to the breadth of content of the current audit. It begins the right sort of conversation, and where ONS can add value is starting to introduce other variables and factors, and starting to broaden out the intersectionality with the data that is available to us.
Q115 Chair: As statistical professionals, are you not worried? We have had evidence given to us to show how much this data could be misinterpreted, I am sure not wilfully, but just by default. For instance, the audit shows that black women are almost twice as likely to have a common mental health problem as white women, but you could also be looking in that same dataset at information like housing tenure and other elements that could provide much more information than that simple fact around mental health, to try to explain why that problem is happening. It may not necessarily be driven by NHS issues or access to NHS issues; it may be driven by a disproportionate probability that they are living in a certain sort of housing or any number of other things, yet that more low‑level data is not readily available.
Iain Bell: The audit was very clear, and the caveats were there quite clearly, about how to use the data. This is always going to be a tension within a statistical system whereby, if you wait for fully understanding and knowing all of the causes of a system, the data may not have seen the light of day for quite a long time. That then means that all the onus is on the people who have the data in order to fully analyse and do all of the causalities. Usefully, the audit, by putting the data together in one place and making it available, can begin to get better questions asked to allow the evidence, analysis and further research to get into explaining the disparities.
I view it as a starting point for the further research and analysis, which enables us to achieve more by having it out there and available to a wider group of researchers than if it was simply waited for until all that analysis was completed by the data holders.
Q116 Chair: You talk about this tension and you are absolutely right. There is a concern, again coming from a different set of witnesses, that if you continue to disaggregate the data you could show that there are almost always other factors at work that would conceal race disparities. What advice would you be giving to the departments currently analysing this data, so that they get that tension right?
Iain Bell: The departments across Government and the people doing this are part of the Government Statistical Service, and they are very used to this tension on a day‑to‑day basis and managing it with the datasets they release. They will constantly be weighing it up, not just with the tension we have there, but also with the risk of disclosure as we get down to smaller and smaller numbers as well. They will be factoring this in.
The best advice we would really be giving is that we are here to help and advise, in that we have a lot of experience on disclosure control as ONS. Melanie put this very well. Fundamentally, it is about making sure that, as a necessary condition for open debate, there is the transparency of the data, alongside the analysis and evidence that they have, where they have it, and the appropriate caveats to make sure that it is fully understood what they can and cannot conclude. That is usually the best way to manage your way through that tension. The transparency of the data is essential.
Chair: That is really helpful. That is great. Thank you so much for your time this morning. It is really helpful in the formulation of our report. I know it takes a lot of energy to come here and to prepare for it, so we are immensely grateful. Thank you very much.
Witnesses: Rt. Hon David Lidington CBE MP and Marcus Bell.
Chair: Good morning. I would like to welcome our Minister, David Lidington, and Marcus Bell, who is here from the Cabinet Office, and two people who have joined us in the audience as well.
Minister, your predecessor in his role was keen, we know, to engage with the Committee about the next steps the Government should take now that the first stage of the race disparity audit has been completed. We are grateful for that, and look forward to assessing the evidence we have heard and putting our conclusions and recommendations into the usual report. However, we want to learn today more about the Government’s own thinking, what you have learnt about the audit and what steps you plan to take to build on that and to respond, importantly, to its findings.
I do not need to ask you to introduce yourselves; I have already done that. I will move immediately on to our first set of questions.
Q117 Jess Phillips: Hello. Your predecessor characterised the Government’s approach to the results of the race disparity audit as “explain or change”. Does this mean that, if a disparity can be explained, the Government do not think that it needs to be changed?
David Lidington: It depends what the explanation is. For example, one line of constructive criticism of some of the statistics that we presented in the audit has been that they have not been adjusted for age, class, income level or other characteristics that might provide a better and more subtle explanation of the phenomena that we are seeing. I would argue that the audit is unprecedented in the sheer scope and scale of what is being presented, and it is backed up by a determination, from the Prime Minister, personally, down, to address racial disparities.
One looks at the evidence. One looks, for example, at the evidence presented in the audit of education attainment at school. You see different average levels of attainment by boys and girls within each ethnicity, and you see some quite stark differences between the relative performance levels, on average, of different ethnic minorities. One needs to think through what lies behind those inequalities and devise a policy response that addresses that.
Q118 Jess Phillips: In that case, even if you could explain it, you would still seek to change it.
David Lidington: Yes. The audit is the starting point. You then say, “Look, if there is this inequality of outcome, is there some reason for it?” That may throw up a reason that goes beyond a simple statement saying that people are judging those from one or another minority group unfairly. There may be other contributions to the explanation.
Then you want to ensure that there is genuine equality of opportunity in things like education, and ensure that all public services are operating in a way that is fair to all citizens of this country.
Q119 Jess Phillips: It seems to me that you want to change it regardless, which is not a bad thing. That is not a criticism. It does not sound like “explain or change”. It sounds like “change or change”.
David Lidington: “Challenge” is the key word there.
Q120 Jess Phillips: That is fine. I am happy for you to change things. How are the Government going about explaining the disparities revealed in the audit?
David Lidington: We are now in the process of getting the Whitehall machinery into action on this. There have been a number of early priorities. One which, given my previous job, I knew something about was the Government’s response to the Lammy report on the experience of black and ethnic minority people in the criminal justice system. When Justice Secretary, I announced the Government response to that towards the end of last year, for example, setting some very demanding targets for both recruitment and development into leadership positions in the prison service.
Q121 Jess Phillips: You are prioritising certain areas.
David Lidington: We are prioritising certain areas. One of the things that I intend to do, having inherited this position now, is to use the inter‑ministerial group on racial disparity, which is supported by the unit of the Cabinet Office that Marcus heads, to ensure that each Secretary of State and department is addressing those things that fall within their areas of responsibility.
Obviously there will need to be a system of prioritisation here; otherwise one will just find that effort and resource are spread too thinly to have a proper understanding and impact.
Q122 Jess Phillips: What, if any, deadline has each Government department been given for explaining the areas of disparity?
David Lidington: We are not at that stage yet. I want to make sure that, when we are 12 months on from the initial publication of the audit, we are able to say that some demonstrable progress has been made across Government. The earliest priorities have been the response to the Lammy report in criminal justice terms, and some work that we are undertaking on employment. We are targeting efforts in about 20 areas of the country that have both a high BAME population and a high gap between employment rates for BAME and white citizens, and trying to get at what lies behind those differences in outcome.
Colleagues in the Department for Education will be announcing fairly soon an externally led review of school exclusions because, again, you look at the stats in the audit and see that there are more black kids being excluded from school, both temporarily and permanently. We need to understand what is driving that and what can be done about it.
Q123 Jess Phillips: As you have said, you need to have priorities because the Cabinet Office is only as big as it is. Will the departments be bringing in additional expertise to help them in explaining the findings of the audit, and, if so, what?
David Lidington: It will be for Secretaries of State and accounting officers in departments to allocate the resources to this work. It is something that the PM has made very clear that she regards as a key aspect of the Government’s social policy programme, but they are going to have to do this within the budgets that were allocated to those departments in the spending settlement. When the Secretaries of State make bids for the next spending review they will be taking this into account.
Q124 Jess Phillips: Will you oversee, for example, as a Cabinet Office what exactly they are doing? What will happen if somebody is doing nothing?
David Lidington: There will be meetings of the inter‑ministerial group. The unit—and Marcus may want to add a bit on this—is talking to colleagues around Whitehall the whole time about what they are doing. One thing I have asked the unit for, since being appointed to this role, is a report on where things now stand, because I want to make sure that departments understand fully that they have a responsibility to follow up on the findings of the audit and put some action plans in place if they have not already done so.
Marcus Bell: All the data on the website comes from departments, so it is not our data, but departments’ data, which they sent to us. They have lots of analysts who understand that data very well and are more than capable of conducting a thorough regression analysis on it, to tease out the kind of issues that you are asking about. They are doing that at the moment. Departments have been asked by us to do a regression analysis on all of their measures that are on the site. We are co‑ordinating how they are doing that to make sure there is a common approach.
Q125 Jess Phillips: In trying to explain it, you have analysts in every department that is relevant. It is specifically on data analysis.
Marcus Bell: That particular aspect is, yes. We are also actively involved on the policy side.
Q126 Jess Phillips: I suppose there is no explanation you would be seeking from field work—what I would call “talking to people”.
Marcus Bell: It is for departments to decide how they do that. We have asked them to make sure that they have done a proper regression analysis on the data on the site.
Q127 Chair: Can I ask for a tiny bit more detail? From what you have said, education and the criminal justice system could well be top priorities in what you are doing, and there could well be clear data on racial disparities there; you talked particularly about exclusion. The evidence that we have been given would suggest that, in other areas, there will be co‑dependencies driving disparity. One example we were given was that black women are twice as likely to have a mental health problem as white women, yet are also more likely to be living in social housing. It may be that housing tenure is the problem, rather than race.
How will you bring that thinking together? Health and housing sit in different Government departments. Is there room for a strategy here so that those who are interested will know what priorities you have decided are most important? We have just heard from EHRC and ONS that data transparency is essential for open debate. Strategic transparency is also essential for good scrutiny. Is there room for a strategy here?
Marcus Bell: There is a data answer to that and a policy answer. The data answer is that, currently, on the site there are in effect 130 or 140 individual measures, and we did not have the time in the first phase of the audit to make linkages between the different data sets, but that is something that we could prioritise for the second phase. It would be interesting to know more about the connections between education, employment, mental health, housing data and so on. That is certainly something that we are discussing with departments.
On the policy side, exactly as you say, some of the issues raised by the audit are classic cross‑cutting Government policy issues, where there are serious dependencies between one area and another, and we are also discussing those with departments, particularly in terms of the connections between education and employment.
David Lidington: On your question, Chair, about an overall strategy, we have no plans at the moment to have a formal cross-Government strategy, but rather would expect departments—taking account of the Prime Minister’s priority here—to work within their respective areas of responsibility, sometimes taking particular initiatives in response to the findings of the audit, and sometimes using the audit findings to influence the development of streams of work that they were doing anyway, to provide for more effective policies.
To take your mental health example, the Secretary of State for Health and Social Care is already preparing a mental health strategy, which will be published later this year. He and his team will be looking to the results of the audit and building that in. Similarly the Wessely review was established towards the end of last year into detention under the Mental Health Act. It is a matter of record that black people in particular are more likely, statistically, to be subject to detention under the Mental Health Act than white people. This is something that will be drawn to Professor Wessely’s attention, and he will no doubt want to take that into account in framing his work.
Q128 Chair: You are making choices with the way you use scarce resources. How can this Committee scrutinise your choices if they are not set out in a strategy?
David Lidington: The choices ultimately in policy terms are ones for Secretaries of State in their own departments, but through the inter‑ministerial group we seek to remind—that word sounds a bit too weak—Secretaries of State that this is something to which the Prime Minister personally attaches a very high priority, and to ensure that they are continuing to work on this.
If it would be helpful, we can look for ways to provide a report to the Committee from time to time on progress, so that the Committee has a view of what is happening across Whitehall. I see no problem about that as a matter of principle, but sometimes the work to address the disparities found in the audit will take place within the context of a broader policy work stream that a particular Secretary of State is taking forward.
Chair: I am sure that the Committee would always relish any information that the Government could furnish it with.
Q129 Tulip Siddiq: Thank you for coming in. I am really surprised by what you have said so far. You have had the data from the audit for a very long time now. Most of the findings are not a secret. There is no doubt that there is a link between people who do not receive a lot of money in terms of employment, with the gap in pay for ethnic minorities, and those who are trying to rent. There is no chance of home ownership.
The approach sounds very piecemeal in what you are saying. Unless you have a cross‑cutting Government strategy, you are not going to be able to solve this, and I am sure that you are aware of that. Is there no chance of having perhaps a Minister for Race who can co‑ordinate everything that is happening across the departments?
If it really is a priority for the Prime Minister and she is attaching importance to it, which we welcome, it does not sound like you have an overall strategy. Correct me if I am wrong, but it sounds like each department is doing its own thing. That will not solve the problem overall.
David Lidington: It is important that each department feels that they have responsibility here. While I can understand Ms Siddiq’s argument, the risk with designating a Minister somewhere—a Minister for Race Equality, to take a description at random—is that departments then say, “It is that Minister’s job to sort this out, rather than it being my job as the Minister for Schools, the Minister for Courts, the Minister for the Prison Service or the Minister for the Police to address these problems in my particular areas of responsibility, and deal with the public services for which I, in my ministerial capacity, am responsible”.
The fact that the unit is located in the Cabinet Office gives us a locus and co‑ordinating role to make sure that departments are being kept up to the mark, but I do not think that it would be right to portray the Government’s response as in the least sluggish. We are determined to make certain that the initiatives we take identify and then apply policy solutions that are going to work.
Let us take Ms Siddiq’s point about employment. What we are doing there is to establish 20 pilot projects, as I described, in areas of the country where we have a high BAME population and a large gap between ethnic minority and white employment rates. We have Jobcentre Plus working with local employers, local authorities, the voluntary sector and others to understand and take action to remedy those gaps, improving the training for jobcentre staff who are dealing with people from minority groups seeking work.
It means talking to local employers about things like having anonymised job applications and addressing possible questions of unconscious bias. It also means talking to training and education providers locally because in some cases we may find—and I do not want to speculate on the results we get from these pilots—that young people from particular groups are, for some reason, not getting access or gaining the qualifications for which employers in that area are looking. We need to get to grips with the reasons at the really concrete level, and then put in place the remedies for them.
Q130 Tulip Siddiq: I do not think that any of this data comes as a surprise to anyone. It has been around for years and years that people from the ethnic minority community are not getting the right kinds of pay or the jobs that they want, but to deny the links with housing is astonishing. If someone does not get paid enough money, cannot get on the housing ladder and cannot rent, you have to address the problems in the rental or housing market. You cannot deal with these departments individually. You must realise that there has to be a cross‑cutting strategy.
I do not think that establishing a Minister for Race would be evading responsibility. It would be attaching importance to an issue that has been discussed for years, but successive Governments have done nothing about.
David Lidington: First of all, I have not ever said that we will ignore the housing question or a possible relationship there. When it comes to employment, there have been some good signs of progress. One of the groups of people who are still least likely to be in work are women from Pakistani and Bangladeshi communities in this country, and it is way below the national average for female employment, but in the last 10 years we have seen a significant increase in the number of women from these communities who are getting into employment. There are success stories about the country that we can learn from and build on as well.
Q131 Tulip Siddiq: Are you satisfied with the level of increase and number of ethnic minority women who have jobs in 2018?
David Lidington: No. The fact that it is way below average means that you cannot be satisfied there.
Q132 Eddie Hughes: Where disparities cannot be explained, what deadlines have been set for departments to come up with a plan for how they will bring about change?
David Lidington: We are still at relatively early stages here, and you will appreciate that I have been in post for coming up to a month now. In terms of the work that has already been done at official level, we have not set firm deadlines for this.
Marcus Bell: As you said earlier, the general aim is to be able to demonstrate very significant progress in tackling some of the key disparities in the course of the year, i.e. by October. Some departments are rather further ahead than others and discussions continue.
Q133 Eddie Hughes: What streams of work have been started in Government in response to the audit’s findings?
David Lidington: Lammy admittedly preceded the audit, but the Government response to Lammy was a significant step forward: the pilot project in jobs, the exclusion review where we will have an external lead, and a panel that will be set up externally to DfE to advise on that.
Then we have a number of work streams that are going on within Government about this. I mentioned the work that was going on in terms of the Wessely review in mental health. If we go back to education, the previous Secretary of State for Education last autumn, shortly before Christmas, launched a social mobility plan for the department. The findings of the audit are being built into the ongoing work of the department on social mobility. In addressing the broader issue of social mobility, it is taking account of the particular challenges thrown up by the race disparity audit, and that is not to the exclusion of everything else. There is a real challenge with kids from poorer white families in terms of underperformance in schools as well, but that is going to be part of the mainstream work of DfE.
Q134 Eddie Hughes: What criteria are you using to determine how you focus your efforts?
David Lidington: We are looking at the findings of the audit. We are saying to departments, “Look, where there is an inequality of outcome, the PM has set this challenge that you change or explain”. We accept that sometimes there will be a cross‑cutting issue here. Other times it may be that you are having to address some quite deep‑seated issues about a relationship between people in a particular community, the operation of a public service and the expectations that one has of the other, and to then come up and come forward with your solutions.
This work is not going to be accomplished in six, or even 12, months. This programme of work will need to be followed through for a period of years and under more than one Parliament, if we are going to get the genuine equality of opportunity that we want to see.
Q135 Eddie Hughes: It feels to a degree that the problem could be so broad that it is very difficult to focus down on specific areas and prioritise. That is the first thing.
Secondly, to what Tulip said, this information has been around for a long time. I appreciate that you have only been in post for a month, but in terms of deadlines, will we—or a different version of this Committee and a different Minister in the future—be able to sit here and say, “Yes, we know it is tough and very broad. What has actually changed?”
David Lidington: The instruction I have given to officials is that I want us to be in a position—and there will be an expectation in Parliament and outside more generally—12 months on from publication of the audit, in October this year, to say in public, “This is the measurable progress that has been made”. When I met some of the leading NGOs in this field a couple of weeks ago to discuss their views on the way forward, they felt that that timescale was a reasonable one too.
Q136 Tulip Siddiq: Organisations like Operation Black Vote have welcomed the audit, but their main question when I have been speaking to them is about what difference this initiative will make with regards to lots of initiatives that have taken place in the past. Do you have any plans to have a serious budget for direct action in targeted areas? Do you have any plans for measurable targets? Are you going to focus on procurement in tackling racial disparities when it comes to employment?
David Lidington: First of all, on OBV and other organisations, like Ms Siddiq, I have found when I have talked to them that they are very welcoming of this. In response to the challenge that they understandably have made, I have said that the difference is that this is going to be a sustained period of work. It is not going to be something that we publish in order to get a headline in October 2017 and are then going to forget about.
That means that we are not going to be looking for glib headlines and something that we can claim is an instant result. It has to be steady work, target by target, priority by priority, to deliver the right outcomes over time. OBV and others have said to me that the fear of the people whom they represent is that it is a flash in the pan. We have to demonstrate sustained effort, more than a result in two or three months’ time.
On budgets, I said earlier that every Government department operates within the budget that it was allocated in the spending review. The work that departments want to do in the next spending period is something that Secretaries of State bid for in the normal way. There is a limited budget available in the Cabinet Office to support cost‑cutting work, but we look to departments, because this is about achieving the PM’s stated social policy objectives, to find money from within their existing resources to deliver the right results here.
Marcus Bell: On scrutiny and follow‑up, there is an important point about the product from the audit not being a report but being a living website. There is already some time series data on it, but as the data changes we will add it so you will be able to see month by month, quarter by quarter, year by year whether things are getting better or worse for each of the measures.
David Lidington: There is a measure of accountability in that.
Marcus Bell: There is some accountability and also pressure for change.
David Lidington: Yes, it is not one set of stats and that is it.
Q137 Tulip Siddiq: When you want to implement the findings of the audit, how will you be measuring the success of the initiatives that you are planning to implement?
Marcus Bell: That is a slightly different issue. We will update data on the website for all the measures as it changes, and quite a lot of it has changed since launch in October, so there have already been a lot of updates. There is a separate question, as and when departments set out their initiatives to tackle disparities, about how we evaluate the impact.
Q138 Tulip Siddiq: Once again, and I know that I am pushing my luck here, do you not think that a cross‑cutting strategy would help you to measure the success of this a lot more than having individual departments doing their own thing?
David Lidington: The gap between what Ms Siddiq is calling for, a cross‑Government strategy, and what I am describing, which is departments building this work into their mainstream business and as part of their key priorities, with a co‑ordinating mechanism provided by Marcus’s unit and the inter‑ministerial group that I chair, is perhaps less than Ms Siddiq fears.
Q139 Tulip Siddiq: Have previous Governments not tried that and failed?
David Lidington: I do not want to pass judgment on previous Governments.
Q140 Tulip Siddiq: What about your own Government before the election?
David Lidington: Ms Siddiq, it is fair enough to say that this is something successive Governments over the years have talked about, and we still see some of these inequalities of outcome.
Q141 Tulip Siddiq: That means the strategy did not work. Previous Government strategies have not fixed it.
David Lidington: There is no magic to putting the label of “cross‑Government strategy” on something. I have seen plenty of those come and go in my time in Parliament. What I believe can make the difference here is the personal commitment of the Prime Minister. The reaction of people in a lot of departments at the operational level—this is the advice I get from the Cabinet Office team—has been very, very positive: that this is something that key officials in Government departments really want to make a difference over.
Q142 Tulip Siddiq: I seem to remember a previous Prime Minister being committed to the big society. I do not know where that one went. Anyway, as an aside, I just want to ask one more question. Is there a reason why the audit did not refer to the public sector equality duty, which I thought was a useful tool for identifying racial disparities?
David Lidington: It simply did a slightly different thing. The public sector equality duty, as the Committee knows, requires public authorities to have due regard to the need to eliminate unlawful discrimination, harassment, victimisation and other things outlawed by the Act, and to advance equality of opportunity and foster good relations.
Those duties apply. Here, we are focusing on the outcomes. What is the experience of people from BAME communities and how can we make sure that public services genuinely are providing equal treatment and equality of opportunity for them? Those services are already under the public sector equality duty and, in my experience, management and leaders in the public service are very well aware of their responsibilities in that regard. We now need to probe why, despite the existence of that duty for some years, there are still the disparities that were shown last October.
Q143 Chair: I just wanted to pick on one particular issue. You talked about departments building racial disparity into their work, which is really welcome. When I think back to the ministerial working group on Gypsy, Roma and Travellers, which was constituted under the coalition Government in 2012—and we have obviously been looking at that piece of work—it promised that Gypsy, Roma and Travellers issues would be included in reviews and reports in the future from 2012 onwards. When the reports were published—and I am thinking here about things like the hate crime review, the GP guide to commissioning, the exclusion review in 2014 or bullying guidance—GRT communities did not feature as had been promised.
How can we make sure that the new reviews do exactly what you, as the Minister, want them to do and take into account the race disparity audit information and the work of the race disparity unit? It seems like there have been good intentions in the past under a Conservative‑led Government, but we want to make sure that this is going to be continued in the future in a way that can be monitored and the Government can be held to account.
David Lidington: We are held to account partly by the various statistical series that Marcus described—the website is a living source of information—and partly through the inter‑ministerial group where the relevant departments will be represented. You are right, Chair; the October stats showed again that Gypsy and Roma children really come bottom of all the tables in terms of school attainment.
Marcus Bell: We work quite closely with organisations representing the GRT community on the audit, and we are keen to include what data we can about Gypsy, Roma and Traveller people, yet there is relatively little on the site. A couple of reasons for that are worth spelling out. One is that some departments do not collect it. They do not ask, “Are you Gypsy, Roma or Traveller?” There is no box to tick. That is one issue and we can only publish the data that departments have.
The second reason is that, even where it is collected, the numbers are small and the distribution is relatively even across the country, so if you get below a very general level of analysis you are talking about very small numbers, and there is relatively little to say. None of that means that going forward there is nothing that can be done. We were keen to include what data we could about Gypsy, Roma and Travellers, but for those specific reasons there is not very much on the site, with the exception of education, where there is quite a lot of detailed information about GRT children.
Q144 Chair: How will you be supplementing the data in the race disparity audit in order to address GRT better in the future?
Marcus Bell: We are talking to departments about what data they collect, about which groups and about which categories they use, with a view to getting a bit more consistency in their approach. Gypsy, Roma and Traveller is one of the standout issues there, because one department, DfE, has an awful lot of rather disquieting information about GRT children and how they are doing. Other departments know relatively little.
David Lidington: October was a start and we are talking to departments, but also listening to what outside bodies are saying about the additional categories of information that they would like to see collected and published on the website. If the Committee has suggestions for that, I am sure that we would want to take those into account.
Q145 Angela Crawley: Turning to the Government’s engagement with civil society, how do you expect the Government to engage with civil society to ensure that they can address the disparities revealed within the audit?
David Lidington: I made a point in the last couple of weeks, at an appointment, of meeting a number of the leaders of some of the key civil society organisations in this field. At official level, that is something that is done very, very regularly. I am not certain how often you talk to the OBV, Runnymede Trust and so on.
Marcus Bell: Extensively and all the time is the answer. We try to make a point of that in terms of how we do this work, engaging with lots of voluntary organisations, but also members of the public.
Q146 Angela Crawley: How do you expect to support civil society to deliver on these outcomes that you are discussing?
David Lidington: It is Government, primarily central Government, that is the priority here. The purpose of the audit is to provide the statistical basis on which to shape policy for the delivery of public services. We are getting some local authorities and NHS trusts coming to us, having seen the audit results, and saying, “We would like to learn from this. What can we do in our areas of responsibility?” Civil society I see as important partners, and it may well be, as for example in the pilot projects on employment, that we conclude that working more effectively with civil society on the delivery of public services, for example on the mentoring of people, is the best way to address the disparities that the audit has revealed.
Q147 Angela Crawley: Coming back to your previous points about the role of the Cabinet Office, you have mentioned how you see the role of the Cabinet Office as co‑ordination and almost oversight. How are the Government departments going to be held to account for their work in response to the audit, and who will be responsible for assessing their plans and evaluating that progress?
David Lidington: It will be through the ministerial group, which will be served by the racial disparity unit. The unit will report to Ministers, to me and ultimately to the Prime Minister on the progress that has been made or where we are finding problems.
Q148 Chair: Moving on to the last couple of questions I have, what has the Cabinet Office learned about the quality of the data on ethnicity that is currently collected across the public sector? Are you really happy, really unhappy or somewhere in between?
David Lidington: I will turn to Marcus for his expertise on this. I am not a statistician, but my take on it would be that it is very varied. Different departments have over the years collected sets of information that they judged were the most important for their departmental priorities. Sometimes that provides, as in the case of education attainment, some quite comprehensive statistics about outcomes for different ethnic groups. Other departments have not previously seen this as central to their role and have not collected the statistics, or the statistics are not easily divisible so that you can come up with meaningful samples for subsets, segmented by age or other social characteristics.
Marcus Bell: That is exactly right. It is clear, if you look through the measures on the site, that there is enormous variation in the quality and depth of data about ethnicity held by government departments. Typically departments where there has been scrutiny and pressure over the years on issues around ethnicity, like DfE, Ministry of Justice and DWP, have detailed, granular and useful data, and others not so much.
Q149 Chair: Is there any Government data that you rejected and would not put on the website because it was too misleading?
Marcus Bell: Yes. We did an initial stocktake with Government departments and that identified about 340 datasets. We have only published 20% of those on the site so far, so potentially there is a lot more to come, but some of them we did not include for quality reasons. That is a point of statistical quality: it does not necessarily mean that it was bad data, although in some cases it was bad data; it is more that it did not satisfy the quality bar that we want to apply to the numbers.
Q150 Chair: When we were talking earlier this morning to ONS about data, consistency of data was something that we raised with it. We are really interested to know what your plans are to improve the consistency of data that is being connected and the ability that you will have to disaggregate that data. Could you spend a few minutes explaining that?
Marcus Bell: We are working with ONS. We have a great interest in consistency in terms of ethnic classifications. We are working with ONS and statisticians across Government to try to introduce more consistency in the classifications. That is very difficult to do overnight, because you cannot suddenly reboot all department systems and ways of collecting data. Slow and incremental work to get more consistency there is important.
In terms of data quality, this was a big cross-Government data project. We are networked into analysts in departments and have had quite extensive conversations with them about how to improve the quality of data, where it is not where it should be.
David Lidington: Is it correct that we try to apply the UK Statistics Authority code in terms of judging the quality of stuff that we could publish on the website?
Marcus Bell: Yes.
Q151 Chair: Do you see consistency and disaggregation as being a priority for the way this dataset develops?
Marcus Bell: Yes. You mention disaggregation. In particular, getting more geographical disaggregation where we can feels important. From the feedback that we have had about the site, some of the more usable and interesting data on it is where outcomes for people with different ethnicities are compared in different geographical areas. That is something that we are keen to do more on.
Q152 Chair: My final point would be that we heard about the work that ONS is doing around the census. It was very interesting to look at how that could be used more between censuses. Is that something that the Government will be supporting?
David Lidington: With the census, there is a clear structure whereby ONS decides what questions are going to be asked. There are obviously a number of demands for different categories to be specified in questions on the census, and those are things that ONS will have to make a judgment about in due course.
Chair: Thank you very much for your time. I appreciate you taking the time out of your very busy diaries to be with us this morning and all the preparation that goes with that. Thank you very much.