final logo red (RGB)


Public Services Committee

Corrected Oral evidence: The role of public services in addressing child vulnerability

Wednesday 16 June 2021

3 pm


Watch the meeting

Members present: Baroness Armstrong of Hill Top (The Chair); Lord Bichard; Lord Bourne of Aberystwyth; Lord Davies of Gower; Lord Filkin; Lord Hogan-Howe; Baroness Pinnock; Baroness Pitkeathley; Baroness Tyler of Enfield; Baroness Wyld; Lord Young of Cookham.

Evidence Session No. 11              Heard in Public              Questions 82 - 89



I: Professor Anna Vignoles, Director, Leverhulme Trust; Dr Polly Vizard, Associate Professorial Research Fellow and Associate Director, Centre for Analysis of Social Exclusion, London School of Economics; Dr Tina Haux, Director, National Centre for Social Research.



  1. This is a corrected transcript of evidence taken in public and webcast on
  2. days of receipt.



Examination of witnesses

Professor Anna Vignoles, Dr Polly Vizard and Dr Tina Haux.

Q82            The Chair: Good afternoon, everyone, and welcome to this slightly shorter session today of the Public Services Committee in the House of Lords, looking at child vulnerability and what lies behind that. We have three academics this afternoon, all of whom have research experience, so we are looking forward to this slightly shorter session. Given the good weather and that several of my colleagues, including me, will probably want to see Wales play later on today, it may be a good thing that it is a shorter session. We welcome everyone who has come to listen and to see what is going on.

First, I will introduce our panel. We have Dr Tina Haux from the National Centre for Social Research, Professor Anna Vignoles, director of the Leverhulme Trust, and Dr Polly Vizard from CASE at the London School of Economics. We welcome all of you. It is great to have you with us this afternoon. I will ask the first question, as ever. When I do, please introduce yourself, so that people can link your face to who you are and what you are saying. That would be really helpful.

The first question is about our definition of child vulnerability. We have been grappling with how different people define vulnerability, and we wondered if you had a view about what policymakers, particularly government, should be taking into account when talking about child vulnerability. What can we all learn, in government but also as parliamentarians, from the latest research evidence on how best to understand vulnerability in children?

Dr Polly Vizard: Good afternoon and thank you for inviting me. I am from the Centre for Analysis of Social Exclusion at the London School of Economics. We are a research centre that does research on, among other things, different dimensions of social disadvantage, looking at the impact of social and public policy.

By way of an introduction, one of the things that we have done over the years is to look at the different ways in which the capability approach, the approach to social evaluation that was developed first by the Nobel Prize-winning economist Amartya Sen, might be extended and applied as a basis for looking at poverty and inequality multi-dimensionallyso looking at living standards and other domains such as education, health, physical safety and security, and different dimensions of individual and social life. We have developed, applied and used that approach in various research projects over the years for equality and human rights monitoring, as well as for thinking about and measuring well-being.

One of the specific projects that we were commissioned to do at one point was to develop a children’s outcomes framework using that approach. That was a research project for the Equality and Human Rights Commission. We were commissioned to conceptualise the framework, to consult on a set of indicators—it was an indicators dashboard approach—and to review the available administrative and social survey data that could be used in an exercise of that type. We contributed to starting to build up that evidence.

It is interesting that you are grappling with the concept and measurement of vulnerability, because that is something that we also very much grappled with and needed to approach, because we needed a way of deciding which children should be separately identified, reported on and monitored within a system of that type. I thought it would be useful for you to know how we went about developing a list of that type.

Our starting point was very much the Commission’s own remit, which was equality and human rights, so we looked at protected characteristics in equalities legislation—disability, ethnicity, et cetera—and some of the intersections of those characteristics. We also wanted to look separately at socioeconomic disadvantage. In addition, we very much wanted to build on the human rights remit and responsibilities of the Commission, and what international human rights bodies were doing and recommending for the development of lists of vulnerable children. This approach was very much supported by children’s rights committees and the Office of the High Commissioner for Human Rights. They have done some work on the development of those lists. There is also a small body of human rights standards and law that relate to vulnerability and vulnerable groups.

We consulted on this approach and our development of a list of vulnerable children, and we suggested that there was a need for, and we recommended, a list of vulnerable children of this type to use for public and social policy purposes. This approach has been used with the Commission’s own framework. I can provide the list in writing, but there are more general lessons that I draw from that exercise and that might be useful to share with you.

The first is that our starting point was very much a universal concept of child well-being and flourishing. Our concept of vulnerability played a role in identifying children at particular risk of harms and deprivations across different critical areas of life, with high risks of abuse, neglect, discrimination and disadvantage. We used the concept of vulnerability to capture the higher risks associated with individual characteristics: family background, additional or different needs or circumstances.

That approach accords with the use of vulnerability in applied ethics, which is a universal concept that people in general are vulnerable. It also identifies childhood in general as a vulnerable life stage. Childhood is a formative and developmental stage with high degrees of dependency in meeting needs and requires special protection, care and love and relationships. Our focus was very much a broad approach to vulnerability rather than one that was very restrictive and focused on one or two particular groups.

Another related broad lesson is that we did not identify a single statutory policy practice definition that would be a good starting point for an overarching approach to child vulnerability, so a cross-governmental approach will inevitably be broad. I have a long list of all the different ways in which vulnerability is used in different social policy areas, both for adults and for children, and it is really wide-ranging, as I am sure you know, with different statutory definitions and thresholds.

Some approaches allow for early intervention and prevention thresholds, whereas others have much higher thresholds. There is use of the term across public services in different ways, in health systems, the children’s social care system, the criminal justice system and so forth. It is really broad. If you are looking for an overarching definition, it necessarily needs to be a broad one that does not invalidate those existing uses and the variation in them.

Another point that will need to be thought through in a definition of vulnerability is exactly the relationship between vulnerability and poverty. We included socioeconomic disadvantage, and you have to address the question of how broad to go. Within children’s social care, there are thresholds that relate to deprivation in basic needs. There is a primary need code that relates to low income, but it has an extremely severe threshold and is not very often filled in. It is to do with asylum-seeking status and not having access to minimum benefits. There are other disadvantage measures, such as free school meals or the pupil premium used in education, as well as food insecurity, destitution and severe child poverty.

If you go as wide as the relative child poverty standard, you are bringing in 4 million children, so you are then going very broad. If you are not going to do that, one point to bear in mind is that a vulnerability strategy needs to be accompanied by a broader child poverty strategy that addresses relative child poverty.

I have two quick further observations. One is that our approach was very much consultative and participatory. We left our list open, so it was not a one-off, top-down, fixed list of vulnerable children; it was open to revision. The pandemic very much changed our perceptions of vulnerability and our understanding of the use of lists of vulnerable children, and the types of children who might be added to that list. Prior to the pandemic, we might not, for example, have thought of digital exclusion as a vulnerability marker.

In various consultations, groups have raised with us whether the concept of vulnerability is stigmatising or stereotyping, so we went over to the term “vulnerable and at-risk” groups; I have seen “very disadvantaged” used as well. That is just something to think through. I am not suggesting that you do not use that term; I am just pointing out that we have sometimes had that feedback.

Professor Anna Vignoles: I am formerly of the University of Cambridge Faculty of Education. I have recently taken up the directorship of the Leverhulme Trust, which is a charity that funds research. Thank you very much for inviting me here today.

I wanted to make one very quick point based on what Polly has just said. Socioeconomic vulnerability, particularly in the context of education, is a major driver of low academic achievement. It is also a major driver of the likelihood of a child having special educational needs. As we will discuss later, the consequences of that for children down the line are very significant. It brings in a large number of children if you use a broad socioeconomic indicator, but it is also true to say that you are bringing them in to your definition for a reason. Socioeconomic vulnerability is a major factor affecting a large number of children in this country, and there is evidence that it profoundly negatively impacts on their development. I did not want that point to be lost.

I would also point out that, when we looked at this with work led by Dr Astle and Dr Bathelt at Cambridge, most of the other dimensions of vulnerability and adversity—factors such as chaotic home life, parental mental and physical ill health and domestic abuse—were all highly correlated with socioeconomic circumstances. It is broad and is a catch-all, but it is there as a major driver of some of the negative outcomes that we would be concerned about.

Dr Tina Haux: I was formerly at the University of Kent and am now director of the Centre for Children and Families at the National Centre for Social Research. Thank you for inviting me.

I wondered whether it might be helpful at this point to talk a bit about how parenting in low-income families translates into some of the outcomes we have been talking about and the link between low income and poor parenting. We did a study a while back on lone parents moving into work. We followed them over three years from when they first moved into eligibility for having to look for work, due to the age of their child, to when they moved into work, and asked them how that changed their family life and their parenting.

A couple of things were very striking. This was before some of the main austerity measures came in, it has to be said, so it feels like it now has historic value in the sense that most parents felt substantially better off moving into work. I am not sure that applies so much now, given the high levels of poverty among lone parents. Coming back to the parenting and home environment that Anna mentioned, they said that being better off meant that they could send their children on school trips. We know that children often self-manage poverty, so they will not pass on invitations to school trips if it involves more money, knowing that their family might not have that money. Participating more broadly by saying, “Now my children can go to the cinema or to their friends’, rather than hanging out in the park”, impacts being better off in terms of the resources and the ability of the children to participate.

That said, Kerris Cooper has done a very interesting piece of research called, Are Poor Parents Poor Parents? Her argument is that, although there is an income-gradient relationship between parenting and behaviour, not just for low-income parents but going right to the top of the income scale, there is a whole range of indicators where it is not so straightforward to say that poor parents are poor parents. In her sample, which came from the big Millennium Cohort Study of 20,000 children, poor parents tended to have more time with their children; they were more likely to spend time reading and doing crafts with them.

I would agree with the vulnerability definition encompassing children in low-income families, and the link between income and home environment is clear. However, by the same token, we need to be very careful that we do not, therefore, automatically associate financially poor parents with being poor parents in terms of how they parent.

Lucinda Platt and I did an analysis of the Millennium Cohort Study, which was a study of children born in the millennium, and looked at how parents’ capacity changed when they separated, and whether the fact that they become lone parents affects how they rate their own parenting capacity. There is often the argument about whether children do worse because they live in lone-parent families or because living in lone-parent families often goes alongside a drop in income. We thought it would be interesting to hear what parents said.

We were able to observe these mothers before and after the break-up. Before the break-up, you could not distinguish the parenting capacity between those who later went on to be lone parents and those who were not. Before the separation, they looked the same. At the point of separation, they would go down quite considerably. A year after separation, as we know from other research, lone parents mental health tend tends to improve and so on, but they never catch up.

As the children in the Millennium Cohort Study got older, all parents thought they were better parents, ironically. Their ratings of their own parenting capacity went up, but the lone parents stayed lower. We were not quite sure how much of this was also due to the quite prominent social stigma around lone parenthood and whether lone parents thought that they were worse parents because that is what everybody tells them, or whether there was something in their parenting behaviour that was different. It is interesting that they think they are worse parents once separated, but that they did not think so before.

Q83            The Chair: That is very interesting, thank you. One of the things that we have become quite concerned about is the number of children who appear to be invisible. There are things happening but nobody picks them up. Have any of you found anything on that in any of your research?

Dr Polly Vizard: We did a project funded by the Nuffield Foundation that looked at missing and invisible children and how we might tackle the phenomenon of data exclusion. We identified some very common reasons why specific groups of children can be invisible or missing from key social statistics: they are often, for example, missed off sampling frames, they might be resident or detained in institutions, or they might live in mobile or temporary accommodation.

Another reason is that their identity or characteristics just might not be recorded in a social survey. The identity of children from Gypsy, Traveller and Roma communities, for example, is sometimes not recorded. Young carers might not be identified, so there might not be social survey questions asking about young carers. Children at risk of abuse and neglect might not be identified. That is to do with the identification of characteristics and identity in social surveys.

The identity or characteristics might be recorded but might not be recognised or revealed by a person responding to a social survey. Young carers, for example, particularly if they are dealing with parent behaviour such as substance or alcohol abuse, might be hidden from view and their role in the family might not be recognised. A parent filling in a survey might not declare them to be a young carer, or their identity might not be revealed because of entrenched discrimination, prejudice or fear, or perceptions of that; the recording of Gypsy, Traveller and Roma identity, when that question goes into a social survey, can be associated with that.

There might be fear of repercussions from filling in a social survey. For example, asylum seekers might be worried about enforcement action. Some of the problem of invisibleness within surveys is because the subject of the survey or the outcome might be particularly concealed or illegal, such as sexual exploitation, child abuse or the tracking of children. Those activities are more likely to be concealed and therefore difficult to pick up.

You have all those problems, and even when identity and characteristics might be recorded, recognised and revealed, sample sizes might still remain very low and too low to support analysis, in which case, as a researcher, you then need to withhold the results for that group because of low numbers and potential privacy issues.

So there is a wide range of reasons why particular groups of children can be missing or invisible within standard social statistics. National monitoring exercises, such as child poverty or child well-being, should address this problem.

In our Nuffield project on missing and invisible children, we had a focus on what we might do about this using existing data. How could we improve some of the headcount evidence on statistically hard-to-reach groups of children and do deep-dive work on outcomes? We built up some new evidence, for example, on children who witnessed domestic violence, young carers, and children from Gypsy, Traveller and Roma communities. I could say a little bit about that if there is time.

The Chair: I have other people wanting to come in, so hold on to that and we may well come back to it.

Professor Anna Vignoles: Very briefly, it depends on what you mean by “invisible” and what you are trying to do about the problem. Another approach is to think about the use of administrative data to identify children who are likely to be at risk, which is a very different thing from what Polly has been talking about, which is using surveys to understand the processes of being at risk. We have made some progress on linking administrative data, because there are multiple datasets that need to be brought together if you are really going to identify vulnerable children with different aspects of vulnerability.

Certainly, the Economic and Social Research Council’s administrative data centre ADR UK has made substantial progress on this, because it is a complicated thing to pull together all the administrative datasets to identify the different measures that might be useful—GP, hospital, HMRC and justice records, et cetera—so that you can pick out children who are particularly vulnerable, but you can build a picture if that data is used correctly.

There are still challenges with that. There is public concern about those linkages, so access to that data is limited, rightly. The cross-departmental co-operation needed to link that data is also challenging. We have the potential to do better in identifying vulnerable children using administrative data, but a lot more work still needs to be done, and we need to take the public with us on the whole data linkage issue.

On top of that, going back to what Tina said earlier, the risk and stigmatisation aspect is important here. If these individual administrative data points are being used to target groups to do things in a general sense, that is one thing; if they are used to identify specific individual children who might be at risk, that is a very different thing, with a lot more potential for error.

Q84            Lord Bourne of Aberystwyth: Following up on the key question about invisibility, we are told by the Children’s Commissioner’s office that 829,000 children are invisible. That seems a very precise figure if they are invisible. You have talked about the reasons why they are invisible, but do you have any thoughts about how we can make them visible? Clearly, that is what is really important. Is it through schools? You mentioned Gypsy, Traveller and Roma children—I am very keen to pursue that; there is a particular issue there—but it is such a significant number that we should be focusing on how we get to those children. I know we are pressed on time, so perhaps we can have some short thoughts.

Dr Polly Vizard: Gypsy, Traveller and Roma children were one of the groups that we managed to build up some new evidence on. We made good use of the innovations in the 2011 census to do that and have some really interesting statistics on the very large proportion of children from these communities who have nobody with a formal education or qualification in their household, and the extent to which they experience multiple disadvantage across different dimensions.

I agree that, as has been said, one of the keys to moving forward with data is administrative and social survey data linkages and admin-to-admin data linkages. We have one new project that will use a new data linkage in this area, which is the Growing Up in England data linkage, which will link the census with other records such as education, FE and, eventually, higher education records. We will be able to pick up many more Gypsy, Traveller and Roma children and young people through that linkage.

I very much agree with Anna that one really important thing that we can do in linking social surveys and admin data is to get from the admin side identifiers of vulnerability, so that, if we do that linkage, we can identify those groups, such as Gypsy, Traveller and Roma communities, through admin sources. Admin sources are really good for sample size and often for vulnerable children, because they are the ones who are often in records, so they do help with identification. The social surveys can provide what they are missing, which is information about family backgrounds, parental circumstances and outcomes.

There are lots of research efforts in this area, as we were just hearing. For example, Understanding Society linked to the National Pupil Database can provide information on life satisfaction by special educational needs and disability status. There is a whole of range of examples of really new and exciting data linkages that are happening. Quite a lot is happening now on children in care and looked-after children, and their education records.

That is the way forward. We need to make sure that the benefits of administrative linkages and the new thinking about data and the power of data, particularly data that has come about as a result of the pandemic, are really harnessed for children. The Growing Up in England dataset came out of a strategic partnership that specifically aimed to build up evidence on vulnerable children, so it is very exciting and a really important development.

Dr Tina Haux: I agree with everything that Polly and Anna have said. The UK is already at the forefront of all this. We are at a point where we have a lot of data. There is the ECHILD database and the SAIL database, on top of Growing Up in England.

At the moment, the Department for Education is commissioning a special study on children who have been adopted or who have special guardianship orders. They want to do a five-year longitudinal study. Trying to get enough of them to have about 500 left at the end of five years means contacting all local authorities to get a sample from them.

Another thing to mention that might be worth knowing is that there is talk about another big cohort study. The Early Life Cohort Feasibility Study will be commissioned in the next couple of years, and UCL is consulting now about particular vulnerable groups that should be included and how to include them in this commissioning work. Hopefully, we will have more data in the near future.

Q85            Lord Young of Cookham: It has been very helpful to get the academic perspective from the panel to complement some of the front-line experience that we have had so far.

The question that I was due to ask, which was about gaps in data, has been covered by the panel, so I wonder if I could come at it from another angle. We have discovered that there are huge ranges of vulnerability, and you have mentioned some this afternoon: poverty, parental split-up, disability, drugs and sexual abuse. At some point, either this committee or the Government will have to make some difficult decisions on priorities.

What has your research told you about where you get the best value for money in addressing particular types of vulnerability? Related to that, are there particular age groups that we should focus on if we are to get the best value for money, given that there will not be enough resources going around to tackle all types of vulnerability for all age groups?

Professor Anna Vignoles: You are absolutely right. As an economist, I am acutely aware that your definition needs to reflect the resource that you have available. The less resource you have, the narrower you want your targeted resource to be. It is likely that targeting a smaller number of children with sufficient resources will be more effective than spreading a limited pot too thinly.

In terms of where one would make an investment, I would like to make two points. First, we need a better answer to that question. When we link administrative data on what we might do to intervene in children’s lives, we then need to link their experiences to their later outcomes, because it is only by reframing the way we think about spending now that we will start to have an answer to your question.

The work of James Heckman shows, for example, that interventions early in a child’s life are likely to be effective and lead to greater long-term benefits, so we really need to be able to link childhood adversity with later outcomes and to quantify those later outcomes particularly in terms of adults’ use of health and justice systems. We can then get a cost-benefit analysis of the costs of early intervention but also the benefits from the savings we might make down the line in terms of the health and justice systems, just as two examples.

Linking that early adversity to longer outcomes, where the outcomes are very much about the use of public services, is where we would get a lot of traction. Professor Terrie Moffitt and her team have been doing this in New Zealand, linking crime, health and social welfare records. They have developed an interesting framework to consider the economic benefits of prevention, so that we stop thinking of it as purely spend and start thinking of it as investment.

Secondly, there is an overwhelming amount of evidence suggesting that early intervention is both key and effective. If I have a pound, where would I spend it? I would spend it early. We know that children turn up to school at a big disadvantage if they are vulnerable in many dimensions, particularly socioeconomically. That is the easy answer to your question.

The harder answer to it is that, if you do a one-shot intervention in early childhood and do not follow up, you are not likely to make a massive difference to those children’s lives. You need to have a framework for investing in children and young people during their schooling, so that they come out at the other end with higher educational achievements and better outcomes. I do not think that we can put all our pounds into the early years, but that would be an area for priority.

Dr Tina Haux: I absolutely agree with all of that. Spending it in the early years is the most crucial part in allowing children not to be behind from the start. I also agree that it needs to be followed up.

The other thing that I would mention is to think about not just the children but their families. You need to think about the £20 increase in universal credit and so on. If the parents are stressed and worried about money, if it affects their mental health, all the support that they are giving to their children will not be able to be fostered at home. So that is the second pound I would spend on the family in order to lift them out of poverty.

Lord Young of Cookham: Polly, do you have a footnote to that?

Dr Polly Vizard: I have already made the case for keeping a framework like this broad. In addition, we need to change the way we think about investment in people. If we can show that there are savings from early interventions of different kinds in all of the children on the Office of the Children’s Commissioner’s list, for example, it would be important to do that. When we think about investing in a bridge, for example, we think of it as investment and separate it from current spend. In terms of post-pandemic recovery, it is important that we have a different approach to investing in people that takes account of long-term savings that can be established.

Also, when we think about whether a vulnerable children’s framework should be more restrictive than that proposed by the Office of the Children’s Commissioner, consider the following challenge: which of the groups of vulnerable children included in that list are not priorities and are not deserving of measures to address their disadvantage?

Q86            Lord Filkin: I would like to ask Anna to develop exactly the space that George started to open up. There will be lots of crisis points of intervention, where the state ought to be intervening, but we are particularly interested in saying something clear and strong to the Government about where they should focus, not least about preventive actions. What you are implying about building the evidence base about which forms of early intervention look as if they are most beneficial in reducing risk, harm or lifelong inadequacies or failures is absolutely fundamental. Could you give us an indication about when you think this will be ready, and a flavour of some of the early evidence about which forms of early intervention look as if they are generally the most powerfully preventive?

Professor Anna Vignoles: There are, in my mind at least, two types of intervention. One is a system intervention and one is a targeted intervention on a particular group of vulnerable children. We have to acknowledge that you are not going to make as much impact as you would hope with a targeted intervention if you are working against a backdrop of a system that is in stress, for want of a better word.

To give you an example, if we think that we want to target children at risk of mental health vulnerabilities, the fact that children’s mental health services in schools are really struggling to cope and are totally overloaded means that a specific intervention is unlikely to be successful, because the broad system is not necessarily working properly.

When it comes to educational achievement, you have a similar thing. If you have a widening gap in earnings inequality between households, whatever you do inside the school system is trying to work against a backdrop that will make it quite difficult to improve children’s outcomes.

The final point, particularly on the education side of things, is that we know that socioeconomic disadvantage is a key driver of children’s outcomes. What really matters is not so much the school that they attend but the teaching quality in particular. Again, although you might be looking for an answer that is about a specific intervention, one of my responses is that, in a system with higher teaching quality, that will benefit not only all children but particularly disadvantaged children. There is good evidence that teaching quality is more important to socioeconomically disadvantaged children, which makes sense, because their parents find it harder to compensate for any lacking in their teaching.

I am just giving you those general points, because I do not think that you can neatly divide prevention into specific interventions and system improvements that are needed. If you want to get into the nuts and bolts of specific interventions, we have some good evidence from the school system. The Education Endowment Foundation has a reasonable evidence base now on what works, if you want to use that phrase: one-to-one tuition, effective feedback to students, peer tutoring, et cetera. These are interventions that not only help all children but are specifically likely to narrow the socioeconomic gap. In terms of what schools might do, we have some good evidence on that.

The big issue, of course, is that what is happening in the family is more important than which school the child goes to, and children are arriving at school with a big disadvantage. The two things that we can do are to reduce poverty, as I have said already, among low socioeconomic-status families and to ensure higher-quality preschool provision. There is lots of evidence from the US, and indeed from the UK now, of early preschool interventions being effective. We could follow up with a list of specific interventions that have been found to be effective, if that is useful.

Lord Filkin: It would be because, although one grasps that reducing socioeconomic inequality is a fundamental platform, it will not be quick, easy or certain. If we are to say anything that will be impactful on government, we no doubt have to make that general point about child poverty, but we also have to make some specific ones about what things one could do that do not take 10 years or so. Anything that you could offer us in that respect would be very valuable.

The Chair: We will come back to that later in the session.

Q87            Lord Hogan-Howe: I am conscious that my question probably overlaps some of the answers already, so if you do not mind I will just adapt and try to build on some of the answers we have heard already. It follows on from Professor Vignoles’ last comments. From many of the witnesses, there seems to be a reluctance to define what vulnerability means, not only generally but specifically in some areas. As Professor Vignoles said, there is reluctance at times to be specific about what might help vulnerable people. Do you have any views on that? To describe it generally is helpful—you do not want to exclude people who might be helped—but it is not helpful in prioritising public resources. Have you formed any views on that? I will start with Professor Vignoles, who made me think about that.

Professor Anna Vignoles: I take your point that being more specific is more helpful for policymakers. In response to Lord Filkin as well, I would just make the point that one of the challenges we have here is that children are a long-term project and, unfortunately, policy tends to be shorter-term, which leads to this tension that we would prefer to do an intervention that is over the next year than to fix some of the more crucial bits of the system that take longer, such as continuing to work on the status and skills of our teachers and early-years professionals to make sure that they are of the highest possible quality.

If you are asking what my take on the evidence would be, it would be that socioeconomic disadvantage, particularly acute socioeconomic disadvantage, will capture a lot of vulnerable children. On intervention, high-quality preschool—the emphasis on it being high quality is crucial—will set those children off in life in a much better way. Being blunt, the greater the earnings and income inequality, the higher the level of poverty and the more children in poverty we have, the more essential that early preschool experience will be if we are not to see a widening of the socioeconomic gap as children progress through the system. I have given you a short list of things that I would see as a priority.

Lord Hogan-Howe: That is helpful. On your final point about income inequality, I think we can all understand why that would be significant when you are talking about socioeconomic factors. Is the basic challenge to get the bottom of the socioeconomic group higher rather than worry about the gap at the beginning? I am not saying that you should never worry about the gap; I am just asking whether at the beginning the idea should be to get the level of poverty reduced.

Professor Anna Vignoles: Yes, absolutely. Living in poverty has a negative impact on children’s outcomes, because their lived experience is dramatically different from the lived experience of a child who does not live in poverty. We can go through the ways in which that works: they have a poorer home-learning environment, their parents are more stressed; their nutrition is not of a high quality, they have less space, they may have less physical activity due to less space, and the parent-child interactions are more fractious because of the environment. It is not that being poor makes you a bad parent, but that being poor causes that home-learning environment and the interactions that you have with your children to be of lower quality.

You are right to say that bringing up the bottom is clearly the key element there, and we as a society have to decide what that minimum looks like for families in order to ensure that the environment in which our children are being raised is at a minimum acceptable standard.

Lord Hogan-Howe: Dr Haux, from all the evidence you have seen and all the research you have absorbed over the years, where do you forensically focus your attention around vulnerability so far as it applies to what Governments can do to help people?

Dr Tina Haux: As well as fully endorsing what Anna said, I would make two points. I can see your frustration that we do not come back with, “This is how we define vulnerability”. Academics are always pushed for narrow definitions, and Polly outlined the complexity in the different groups. Listening to you, I was also thinking, “Vulnerability to what?” A lot of us often focus on economic outcomes, earnings later in life, education, and cognitive and socio-emotive behaviour, but there is vulnerability to harm. Perhaps for you to be clear, the question to ask is what you think is the most pressing vulnerability. I do not think I would be able to decide.

Coming back to the schooling point and the fact that children are behind by the age of three, when they get to where most of the formal childcare is starting in this country—although, for poorer children, there is support now starting from two—I wonder whether it is worth us not thinking about this as a new problem and therefore not thinking that we need new policy solutions. It is certainly something that, in the 1990s, was very much at the forefront for policymakers, because we then looked at the long-term effects. We had interventions that were perhaps unhelpfully adapted. What was Sure Start was really based on Head Start, the US reading programme, which was quite different from what Sure Start became. Head Start was a very targeted programme focused on helping people with the lowest income to read with their children, and it was quite positively evaluated.

At the moment, there is a whole set of new programmes, including A Better Start, funded by the National Lottery Community Fund and focusing on young children. We are still working on and doing more research on this, but, if we look internationally, we have a lot of solutions and policies already that we could do now if we wanted to focus on closing the education gap, for example.

Lord Hogan-Howe: This is my classification, so it is probably wrong, but there seems to be potential for there being chronic vulnerability, where there is low socioeconomic activity generically for a family or a particular group, and there might be times when any one of us is vulnerable, for example if a family breaks up. Professor Vignoles talked about separation being one critical event, but is there any other evidence about acute versus chronic vulnerability?

Professor Anna Vignoles: Going back to the point that this is not a new problem, the troubled families programme did exactly what you have just suggested: instead of taking a broad definition, or multiple definitions, of vulnerability, it identified children at risk of cumulative and multiple vulnerabilities. It was incredibly targeted and by no means perfect, but it had two good features: first, it was found to be effective in places; secondly, it used data very effectively to try to evaluate the programme and improve it as it went.

I would say that this is not new, and that is an example of a programme trying to do exactly what you said, which is to implement a workable but flexible measure of acute vulnerability in a way that you could deliver resources. That builds on what Tina was saying about the importance of recognising the difference between something that is targeted and something that is broad.

Q88            Lord Bichard: Some of the questions I was going to ask have just been asked, but perhaps as a post-script I will ask all three of our witnesses whether there are any specific evidence-based interventions that they have not touched upon today and that we should be considering in this country to tackle this issue of child vulnerability. Maybe they are issues that impact on the parent as well as the child. Are there any specific evidence-based interventions that we have not talked about to date?

There is quite a lot of talk at the moment, following Dame Andrea Leadsom’s report about family hubs, which look alarmingly or impressively like Sure Start. What does the evidence tell us about the likely impact of family hubs? Are there lessons that the Government can learn, from here and abroad, about how to make those even more effective?

Dr Polly Vizard: I am happy to flag up one study from the LSE’s Care Policy and Evaluation Centre, which underpins some of the work from the Department of Health and Social Care on the long-term economic case for investment in mental health. It provides very precise information about economic pay-off per pound of interventions for adults and children, but the children’s ones might be of most interest to you. Those are interventions for social and emotional learning programmes to address conduct disorder, interventions to reduce bullying, health visiting and reducing postnatal depression. There is economic modelling of 15 different interventions with some very precise costings and benefits of that investment.

On family hubs, I do not know very much specifically about them, but the Sure Start programme has been massively de-invested in over the last decade. We have another research programme funded by Nuffield Foundation that looks at the sums involved. We have heard that early-years intervention is very important, and I would flag up that adolescents also need support. One of the positive developments from the last period has been preventive work on violence, for example, some of which is with teenagers. That can be cost-saving too. It is very important to pick up on older children and young adults as well early years in terms of a human investment approach.

The IFS has brought out a very useful new evaluation of the Sure Start programme. I do not know if you have that, but it shows, using very good evidence, benefits in reductions in hospitalisations. It also produced an estimate of the savings that the programme generated as a result of reduced hospitalisation. They thought that you could recoup about 6% of the costs of the overall programme just through that.

Lord Bichard: You mentioned mental health. Of course, the Government have just announced some increased investment in the mental health teams in schools. Could I encourage you to take a view on whether that is sufficient to achieve the kind of objectives that you have in mind?

Dr Polly Vizard: I very much endorse what Anna was saying about a two-track approach to intervention; we need both universal and targeted approaches side by side. Universal, whole school-based interventions are really important. They have not been completely rolled out, but it is a really positive development that they were taken forward over the last five years. They need to be rolled out into all schools, and there are absolutely huge demands on children’s mental health services. Child mental ill health was already increasing prior to the pandemic, and we know that there is evidence that shows that it has deteriorated during the pandemic, so we certainly need to upscale efforts. This is one area of public services that was under tremendous pressure before the pandemic, so we need to invest more.

Dr Tina Haux: From my understanding, family hubs do look very much like Sure Start. I am not sure that they have the amount of funding that Sure Start had, which will make it very difficult. It is also worth learning some of the lessons from Sure Start. One of the things that Sure Start particularly struggled with was getting the most vulnerable parents, which used up a lot of their resources and time. There is a danger of thinking that we have a new answer and a new policy initiative and then forgetting the lessons from the past. I very much welcome family hubs—it is a good idea to bring parents to one place—but it will need to come with substantial funding to get exactly the parents who we are talking about.

On a practical note, I know you are looking for things that you can tell the Government to do that perhaps do not even cost that much money. One thing I have been wondering about for a while is why we do not talk about parenting when we have antenatal classes. You have parents captured there, and they come to you. Increasingly, fathers are attending; whenever we talk about parents, we need to be very careful that we talk about mothers and fathers. You have them captured, but we do not talk about parenting then.

Having three or four sessions on parenting would be really useful. One of the sessions should include conflict when you have a child and how it changes, particularly for the first more than others. Having a child will put pressure on the relationship. We see a lot of separations within three years of having a child. Giving parents a realistic expectation and perhaps some tools around how to deal with this would be very useful. We have those programmes. We have Triple P, Incredible Years and a whole range of more specific parenting programmes, such as by the Anna Freud Centre. I know that Triple P and Incredible Years have issues, but we have them evaluated.

Coming back to what Anna said about the broad versus the specific, we do not offer them very much at the moment. At this point, a lot of parents would very much like to take up parenting programmes across the age range. We have just realised how important we are to our children and have had to make up for missing school and so on. That would be the intervention with my pound.

Professor Anna Vignoles: On the family hubs, I am not sure that there is much to add, except that they do look like Sure Start, which is a good thing. Sure Start suffered from the difficulty, as Tina just said, of remaining targeted. If you were going to give advice, given that the budget for them is considerably smaller than for Sure Start, the advice would be, first, to make sure that they evaluate them properly, and, secondly, to make sure that we find sensible mechanisms to target them on the neediest families.

You asked about other interventions. The principle of helping parents parent from a very early age runs through all the evidence that you have heard today. If I was picking out well-evaluated programmes on which there is good evidence, I would include things like the Family Nurse Partnership or Incredible Years. Lots of them are targeted or at least involve parents in a significant way. The parent angle is incredibly important.

I would like to end with a comment on the question you asked Polly about mental health and the additional mental health provision in schools. First, mental health professionals in schools do an amazing job, so this is not meant as a criticism in any way, but if you doubled their number and did nothing differently, you would just reduce the queue of students trying to see them. That strategy will not really help us in a significant way. How you deploy professionals in schools is key.

We have a situation where you are trying to have two routes. You are trying to have interventions that help the vast majority of students to improve their mental health, which has worsened during the pandemic, and help them get strategies to cope with aspects of their lives. Separately, you are also trying to identify seriously vulnerable, at-risk children who have acute mental health issues. The danger at the moment is that they will be missed, because other children are also in that queue and it is very difficult to distinguish between them in a school setting.

Again, that programme could be very impactful, but it has to be doing something that really helps us to move forward on this issue and starts to de-medicalise, if I may use that term, some of the mental health issues that are currently being batted back and forth between schools and children’s mental health services.

Lord Bichard: Do we do enough to help teachers to identify some of the signs of mental health issues? Should the mental health support teams spend more of their time enabling teachers to do that?

Professor Anna Vignoles: I do not think that the issue is identification by teachers, in the sense that we have more children identified with needs than we have resource to help them, so doing more on identification is tricky. You want to identify the most vulnerable children, and some of the other things we have been talking about, such as using data to identify genuinely seriously at-risk children, might be a good way to do that.

In schools you can start to think about interventions. I am not a psychologist, but I can advise you who you might want to speak to. In work done by people like Tamsin Ford at Cambridge, for example, they are trying to do more holistic and whole-school-type interventions to help all children manage their mental health or improve their mental well-being in other ways. That is how you take the pressure of the system, so that you can then have medical professionals identifying the students who are acutely vulnerable, rather than teachers, and then getting them the help they need.

Q89            Lord Hogan-Howe: Something that Anna just said really chimed with something I want to ask. People do not seem to talk much about GPs, who, it seems to me, are a useful source of data in many ways: they go into people’s homes, they have personal conversations with families, and they should get all the data from A&E departments when there are acute admissions. Is there anything about that in the evidence that is around?

Professor Anna Vignoles: On the data side of things, GP records were something I mentioned when I talked about trying to pull in various administrative data pieces of evidence that you piece together to identify either groups of kids at risk or specific individuals. You are absolutely right that there is a lot of information in GP records, so they could helpfully be used to identify children who might be at risk.

Lord Hogan-Howe: That is a fair point. I am sorry to interrupt, but I was thinking that the GP could take the lead rather than for others to request the data. I just wondered whether more could be done, busy as they are, of course.

Professor Anna Vignoles: You would need to get someone who is an expert on the NHS to get a proper answer as to what a GP can and cannot do. From a school’s perspective, when you refer children to children’s mental health services, you do not do that directly. You often do it through the GP, and it is just as congested in the GP area as it is in children’s mental health services. From the school’s perspective, it does not feel like there is a lot of capacity there to lead or help.

The Chair: Thank you very much indeed to our three witnesses. We sometimes strayed from the questions we had agreed to ask you, so I hope you are okay with that. Please send to us any written information that you think we ought to have a look at and have access to, or anything that you think we missed asking you. We are always grateful to get that. We are very grateful to you for covering the range of academic work that you have, as well as the range of research and the challenges that there are. We understand that there are real challenges in collecting data around children. Thank you enormously for your contribution this afternoon.