Welcome to the Treasury Committee on Wednesday 29 October 2025. We are here to discuss child poverty ahead of the Budget on 26 November. I am delighted to welcome Dr Jonathan Cribb, deputy director and head of retirement, savings and ageing at the Institute for Fiscal Studies. He is joined by Professor Kitty Stewart, professor of social policy at the London School of Economics, and Sophie Howes, head of policy at the Child Poverty Action Group. Thank you all for coming. I should declare for the record that Professor Stewart and I are acquainted because she is a constituent of mine.

I would like to declare that I have a family member who works for Save the Children.

Obviously, child poverty has been discussed a lot, and one of the reasons why we are talking about it is that there may be changes at the Budget. Future trends in child poverty are being discussed. That is one of the reasons why it is high up on the agenda. What is your assessment of the trends in child poverty, given the current situation, and do you have any concerns about that?

Projecting these things is always a little bit difficult. We do not undertake official particular projects on this. In general, you would think that poverty and child poverty would rise over time if benefits rise less quickly than the earnings of workers. People who do these kinds of projections will generally project gradual increases in poverty. I would focus on groups of concern, so larger families, because they are more reliant on benefits than income from earnings, and private renters, because the local housing allowance rates are frozen at the moment.

I would go with the projections from the Resolution Foundation, which I think are very consistent with what the IFS has put out. Child poverty has risen since 2012-13; that was when we saw the bottom rate at about 27%. It is now up to about 31%, and the Resolution Foundation’s projection is that it will be 34% by the end of this Parliament, which would take us right back to the mid-1990s before the last Labour Government took office. I emphasise that the increase is entirely concentrated among children living in families with three or more children. Their poverty risk has risen from about a third to about half, and the projection is that it will rise further.

As you asked, I will not repeat what my fellow experts have said, but it is helpful to understand why low-income families are so reliant on social security and why cuts to it have had a real impact on that group. Low-income families are reliant on employment and social security to get by. It is a basic point, but when thinking about child poverty, it is important to remember that wages do not adjust for family size. Low-income families are very dependent on the social security system to meet basic living costs, so they are particularly vulnerable when that system is stripped back. My colleagues talked about what we can expect to see and the projected rises. That links to the two-child limit, which I am sure we will get on to, but it is also worth reflecting on what has happened since 2010. At CPAG, we estimate that £50 billion a year has been taken out of the social security system compared with 2010 spending levels. That is a huge amount. If you correlate that with rising child poverty, you can see the very clear link between child poverty and social security spending.

You said that it is difficult to make projections. One of the challenges is our very low birth rate. I think it is at 2.1 only in Luton, and it is below 1 in the City of London. The City of London is perhaps an outlier; nevertheless the birth rate is not at the level it needs to be. That will obviously have an impact on statistics. How do you account for that when you are making child poverty projections, because if you have fewer children, you have smaller numbers? Obviously, you are looking at percentages. Would you talk us through how that has an impact on these numbers?

It is important that we look at both percentages and numbers, because if you just look at numbers, you might end up showing that the numbers are going to fall, and that is just because there are fewer children. That is why it is really important to look at both.

Obviously, different demographics around birth rate as well. That is the average. Are there certain demographic groups that are more likely to have children in poverty—perhaps groups with larger families, as have you already indicated?

There are significant differences by ethnicity. Some groups are more likely than others to have larger families, with three or more children. That is one reason why the two-child limit is impacting some groups disproportionately. That is one reason that we are concerned. I can give you some numbers on that if it is useful.

That would be helpful.

I do not have numbers here, but generally Pakistani and Bangladeshi families are poorer and more likely to be affected by the two-child limit, because they are more likely to have three or more children. On average, they have higher fertility. I do not know whether their fertility trends have tracked with other ethnicities as fertility has come down.

Fertility is coming down in most groups, but some groups are still more likely to have a third child than others. Children of black, mixed or multiple ethnicities are over three times as likely to be affected by the two-child limit as children from white families. Children from Asian families are more than twice as likely to be affected. We need to remember that children from those families are already at higher risk of poverty because of other structural factors to do with employment, pay and so on, which exacerbate the risks. I should also say that 75% of children affected by the two-child limit are in white families. We need to remember that there is a difference between the composition and the risk, but it is increasing.

I was going to come to that, because a high percentage of people are in work—something like over 50% of people in receipt of universal credit are in work—so a lot of benefits go to people in work. To go back to the child poverty strategy, whose launch the Government delayed from before the summer—we are still waiting to see it—where do you think it should focus, given that such a high percentage of children in poverty have a parent in work? Is that relevant?

Employment can play a critical role in the strategy. At CPAG, we say that the child poverty strategy needs to focus squarely on increasing family income. There are two primary ways to do that: through employment and through social security. It is important to highlight that many of the gains from parents stepping into employment have already been made. On lone parent employment in particular, huge strides have been made—Kitty will probably have more figures—but it is important to understand the barriers that particular groups face in increasing their employment or their wages. Looking at the composition of low-income families, many who are not working face high barriers to work: they might have young children at home, or be disabled themselves or looking after a disabled child. All those are factors at play. When we have high employment barriers and—I made this basic point at the beginning—wages not adjusting for family size, we see that critical role of social security stepping in to top up incomes alongside employment.

Should the child poverty strategy be doing anything specific on work to increase the incomes of low-income households?

More can definitely be done to invest in tailored employment support for parents. We would like to see that as part of some of the work that DWP is doing. It has been a focus of the Department, which is reassuring, but I think it is important for the Committee to understand that that will get only some gains for child poverty reduction. We could spend billions and billions of pounds on employment support and not move the dial much on child poverty. If we talk about, say, 50,000 to 100,000 families or parents moving into work, the correlating child poverty reductions are quite small. Were we to look at spending £2 billion by scrapping the two-child limit, that would lift 350,000 children out of poverty. When looking at what is most cost-effective, you need to consider that.

There is cost-effective in the short term, but then there is the long term. They should both be considered.

To echo that, we certainly want to see employment support and childcare improvements—fixing some of the problems with the childcare system, such as up-front costs and so on. Work has been done on them, but they still present big barriers. It is important to realise that this is expensive; it is not a cheap fix. Sometimes people think, “We can just get people into work, and that does this for free,” but that is not the case. The second thing is although employment plays an important role, it will be a relatively small part of the strategy, for various reasons that we can go into. It needs to operate with the social security system. One thing I would say about how to make work work for families is that many families in work—even when parents are working full-time—need support from the social security system through child benefits, because wages do not adjust, as Sophie said. We have known that for a long time, for a century or more. That is why we introduced child benefits in the first place and why family allowances were introduced. We have just done some new work that shows, first, that it is much harder for a larger family to escape poverty through work, even when an adult moves into full-time work. They are much less likely to be lifted over the poverty line than a family with one or two children, because their needs are greater and that is not adjusted through work. The other thing is that over the last decade it has become much harder for those families to escape poverty through work. If an adult with three children moved into full-time work a decade ago, some 60% of those families would be lifted out of poverty, and now it is more like 30% to 40%, and that is because of the loss of social security benefits. This is not a case of benefits are for non-working people and work is for working people. You often need the social security system even if you are working, and the two-child limit is affecting many working families.

I would also highlight the role of housing costs. Why are people moving into work and not getting out of poverty? In part, it might be how social security is supporting them, but there are also housing costs. We did a study looking at changes in in-work poverty from the mid-1990s up to the pandemic, and the biggest single factor in why in-work poverty had risen—not the level, but how it had risen—was increased housing costs that were particularly hitting middle income and poorer families. Different people will want to use different policy levers to address housing costs, but I think there is an additional issue there on top of work and social security.

There is a lot to dig through there. I am going to pass to John Grady MP to continue.

I have a question for Professor Stewart. Do we have reliable evidence on the quantifiable consequences of child poverty for the individual child’s education, employment and health outcomes?

Yes, we do. We always want more evidence, but there is a growing body of rigorous research that looks at the causal effect of child poverty on children’s outcomes.

What does it tell us?

To summarise, it tells us that poverty is not the whole story. The low income is not the whole story of why children who grow up in low-income households do less well, but it is a significant part of the story. We did some estimates based on a systematic review, and our rough calculations showed that about half of the differences between children growing up in poverty and those not could be explained by the low income itself. There are lots of robust studies that show that when families on low income receive more support and more income, it makes a difference to children’s educational outcomes, aspects of their health and their social and behavioural outcomes. We have quite good stories, quite good research, digging into why that might be. It is partly because if you give them more money such families spend money on their children—on the things that children need, from basic needs to supporting education and so on. It is also very important for maternal mental health and depression, which have a knock-on effect on children.

So you see low incomes as a significant cause of child poverty and therefore the effects on children’s education, employment and health outcomes later on. How does the evidence you have referred to affect the case for Government payments to families to reduce poverty and improve those outcomes?

It makes a very strong case that this is something the Government should do. There is the moral case that we are in a rich country and children should enjoy their childhood—they only get one chance at it—and it is not fair that so many children are going without their basic needs being met and are unable to enjoy their childhood. Also, the evidence tells us that we are wasting talent by not investing in our children. There is a strong investment case. We know there are immediate knock-on effects and costs for public services that have to pick up the pieces. There are good estimates of the long-term costs and what we are losing in terms of these children’s lost potential and future earnings and so on.

I am very lucky to be the MP for a seat with wonderful children, in Glasgow, but unfortunately it is blighted with some of the highest child poverty in the United Kingdom. On those individual children, Ms Howes, do you have anything to add to what Professor Stewart has said?

The evidence is really clear. Kitty has cited a lot of it, but particularly when you look at some of the more qualitative research that both Kitty and colleagues have conducted, and at the series of interviews that Nesta and I have hosted with parents affected by the two-child limit, they talk really clearly about reduced access to formal childcare and extracurricular activities and about reduced opportunities to play and to socialise. I think we can all identify that those things do have an impact on children’s development and education. I thought it might be helpful to read a quote from a series of interviews that we did with families affected by the two-child limit. This is a couple with three children; it is a working household, and the father is in full-time work. The mother says: “As a family we have started to rely on food banks. Winter months are horrendous as we can’t afford a big gas bill so we have to bundle the children with extra blankets. We have just moved to a property with no carpet and there is no possible way of affording any carpet. We’re limited to two loads of washing a week as we just can’t afford to top up our electric meter. The adults of the house…skip meals so the children have full stomachs. My mental health has deteriorated greatly due to the guilt of not being able to” provide “for the family. My oldest is 11 and I’ve never been” on holiday with him. “My partner is hardly at home because he works so hard to keep us afloat so he is missing so much of his children’s upbringing.” I think we can all see that many aspects of that quote relate to lots of different things that are really important for children and childhood; and that is the day-to-day experience of families.

Just a quick yes or no on this, Ms Howes, if you would be so kind. I take it your position is that there is a strong case for payments to families by the Government to alleviate child poverty.

Yes.

Dr Cribb, have you anything to add to what Professor Stewart and Ms Howes have said?

I want to very slightly push back, based on my reading of the evidence. There are certainly papers that show that income can have an important impact in those contexts, and does have an important impact on child development and on education. I don’t think it’s unanimous. My colleagues and colleagues at Nesta have looked in particular at the effects of the two-child limit, which generates a big change in income, depending on exactly when the third child is born—basically before or after April 2017. They have not found any impact of that change in income on school readiness; that is measured at age five in reception. They have looked at various sub-groups and they cannot see any impacts there. This does not mean to say that there would not be impacts at later ages. It does not mean to say there would not be other impacts on health. There are of course impacts on poverty, which we can go into. In the main, I agree with my colleagues here. I just do not speak with as much certainty about quite the state of knowledge.

Am I taking from you that you would say it is more likely than not that low income causes child poverty, or contributes to it, and leads to poorer outcomes, but there is some evidence that points the other way? Is that what you are saying?

Low income absolutely causes child poverty; in a sense, it is definitional. That is important in and of itself. I think the reading of the literature is slightly more mixed about the extent. Of course, small to medium changes in income for poor children lead to quantifiable effects on educational performance. I think there are some studies where it does, and some studies where it does not. I know that is potentially not the most helpful thing to say, but I just do not think it is unanimous that we know that changes in income are going to have big effects on education.

The last question from me is to ask Professor Stewart to give a wee response to that.

I think we do know; it may not show up in every study. I have done a systematic review, and the reason why it is important to look across the evidence is to get the big picture right. It is absolutely true that not every study finds an effect. My systematic review and a more recent one in the Journal of Economic Perspectives, I think—a US one—both reached the same strong conclusion that the evidence points strongly to the fact that there are effects, including on education, when you look at the body of evidence as a whole. We both find that there is a variation of effect, and the sort of money that comes through something like the two-child limit is exactly the sort of support that other studies—money makes more difference to low-income households, and it makes more difference when it is consistent and reliable for families over time. The new IFS result, with the EYFS, is surprising in the scope of that literature. It is worth thinking about why we are not picking up an effect, and I think there are a number of possible reasons why. One is the measure itself: whether you reach a good level of development. I do not want to get too much into the methodology, but the group of children that you are looking at is from larger families, living on low incomes, who are younger in the year because of the way in which you have to identify them. All those things point to a minority—something like 30% to 40%—of those children reaching a good level of development. You are taking children at the bottom of the distribution, making them more disadvantaged and then asking, “Are you doing even worse than you were?” I would expect to see that showing up impacts, but I think we need to keep in mind what we are looking at: a blunt measure at age five. We also need to remember that this is a heavily covid-affected year group; they turned five in April 2022, so that is another advantage on top. They should have started pre-school in September 2020, and many of them will have missed out. Finally, we know that parents do their best to protect their children. Certainly from our qualitative research, when we talk to parents, it is easier for them to do that when their children are young. When we ask parents, “How are your children affected?”, the parents of young children tend to say, “They do not know about it. I am trying to protect them from those impacts.” With older children, it is much more difficult. They are going to school, and want to join the football club or are not able to get a haircut, so they feel embarrassed about going to school. There are all sorts of ways in which I would expect these things to get worse further on. We need to look at the big picture, and we need to be a bit careful.

We also need to make sure that the Government have proper evaluation in place for any of the measures that they may or may not introduce.

Absolutely—that is important.

I am delighted to welcome Debbie Abrahams MP, the Chair of the Work and Pensions Committee, who is guesting here today.

I want to follow up from John. First, on the evidence of the impact of child poverty on NEETs, I have a preprint of some work from the UK Millennium Cohort Study, which shows quite a strong impact: you are five times more likely to be NEET if you have experienced prolonged child poverty. I think that also reflects international evidence. Do you have similar evidence? Would you like to comment on that? I think the other aspect relates to family adversity.

I am not sure I know that particular study, but it is certainly consistent with what we would expect. We see that child poverty is not only correlated but causally correlated with children’s ability to do well in school. I would need to do look at the details, as some of what you are seeing there may reflect other factors about those children’s households, families and the education they are able to get, but it seems very likely that part of it is the impact of child poverty coming through.

Is there anything internationally that you have come across that says similar stuff?

On NEETs specifically, I do not think so—I will have a look and think about that.

In some senses, these differences based on where you grow up are quite shocking, such as the family situation in which you grow up. As Kitty said, that is going to include a range of different things. The thing I would point out is that there is evidence to show that being out of work, or struggling to find appropriate work or training, in that adolescent period is really bad for you over a longer period as well. There are scarring effects that people experience.

But specifically on the impact of child poverty.

Why do we care about this? We care about it because it is poor children, and because people are out of work. My point is that an additional reason to care is the long-term impacts of NEET status.

On the NEETs that we have, for example—this is very much in the public domain and is a source of interest—a preprint from large datasets that is currently being peer reviewed shows that people are five times more likely to be NEET, and that more than half of NEETs currently fall into the category of having experienced child poverty. Could I move on, and turn it on its head? I have one paper here from last year on the impacts of child poverty reduction targets and how those would impact on public health and health inequalities. We know that child poverty impacts on infant mortality: for every additional 1% of children living in poverty, we know that there will be nearly six additional infant deaths—children who will not reach their first birthday—per 100,000 live births. I have some scenarios here that I will read out. I do not know if you have similar evidence. Taking different scenarios around targets for child poverty, the suggestion is that the high ambition reduction in child poverty—reducing it by 25%—would decrease the total number of cases of infant mortality by 293. That is from now, really, to 2033. It would decrease the number of looked-after children by nearly 5,000, nutritional anaemia by nearly 500, and emergency admissions by—wait for it—nearly 33,000. That is very significant. It also has a regional impact. I wondered whether there is something we need to recognise around setting targets for reducing child poverty and the impact that that will have, not just, obviously, in the crucial area of keeping our children safe and enabling them to thrive, but also in terms of public spending on the NHS and social services.

CPAG and the wider anti-poverty sector is strongly supportive of the idea of the inclusion of child poverty targets in the child poverty strategy. The Government themselves have set out that they envisage the strategy being a 10-year strategy. We recognise that we will need to make incremental progress in reducing child poverty, because there is lots of work to do, and having a timeframe like that will allow us to do some really good stuff, and then see the knock-on impact on wider children’s outcomes and everything my colleagues have talked about in the education space, but also thinking about health. Specifically on health, we did a small-scale piece of research, a survey, over the summer with 371 paediatricians at the Royal College of Paediatrics and Child Health, asking them how child poverty manifests in their working life. They were really clear: 99% reported that poverty is contributing to ill health among the children they treat, and 96% said that poor housing conditions were affecting the health of the children they treat. To give a couple of short quotes, one paediatrician said, “We have had children on chemotherapy die because they have caught fungal infections from damp housing.” Another quote is, “I have written more letters to housing providers to ask them to address damp and mould urgently for children with chronic respiratory conditions [in the last two years] than in the previous 18 years of my consultant career.” So, yes, to concur with what you said, there is lots and lots of evidence out there to support that.

Professor Stewart, it is sometimes said that policies to reduce child poverty pay for themselves. As an example, I think particularly of the work of Nobel laureate James Heckman, who modelled out the economic metrics of early intervention and how they are exponentially more effective. Can you justify some of these interventions on financial grounds alone?

Yes, I think we can. I wish we had accounting systems that allowed us to consider investment in child benefits on the investment side of the spreadsheet, rather than just as day-to-day spending. The evidence is strong that, by spending money on this area and reducing child poverty, you bring down costs elsewhere, both in the immediate term and in the longer term. We know that, immediately, we are spending a lot of money on other services as a result of child poverty—education, health, children’s social care. Spending on all those things is pushed up by child poverty. Donald Hirsch has done numbers on that, and it is high numbers of billions. His estimate is that we are spending something like £20 billion extra a year on public services as a consequence of child poverty. Projecting forwards, given that there is fairly good evidence that children are not fulfilling their potential and are less likely to grow up to be higher-rate taxpayers and more likely to need support of their own in the future, he thinks there is £12.3 billion in lost earnings for those adults, and another £7 billion lost to the Exchequer, in terms of lost tax and higher benefits. What he does not do is then project what we are spending on health because adults grew up in poverty and have worse health now. So that is probably a conservative estimate, but we are talking about over £40 billion.

My colleague Helen Hayes MP, who is Chair of the Education Committee, has a particular interest in this area. Under today’s SEND system, an 11-year-old may end up leaving school for perhaps two years because they do not get support until they are expelled, and there are 1,800 children in Surrey in that situation right now. That almost, by definition, wrecks their life chances. It is often family-breaking. One parent frequently quits work to look after their child, which has a massive impact on their career. Quite often that child will end up in social services in their 20s, instead of in employment. Is there modelling out there comparing the overall cost of that with the model we were speaking about earlier—which may be more expensive at the outset—based on early intervention and effective support going forward?

The overall cost of investing in the SEND system? Compared to the overall—

The overall cost of what it is right now.

There may be, but I could not tell you. Of course, there is an overlap there for many children. This is not the same as child poverty, clearly, but there will be an overlap.

Where I am going with this is, should and could the Treasury be making the case that if you intervened early, that would reduce debt to GDP over the long run? Therefore, one should be going to the OBR and taking into account not just the up-front costs of these interventions but the impact on future GDP, and also going out to markets in terms of borrowing?

I think one could make a very strong case for that. It is a bit of a no-brainer that investing in children is a good idea in the long run. Why that does not translate into policy is, I guess, to do with the political cycle and discounting of the future. But, yes, I think you can make a strong case for it.

There is a good case to be made that providing more money for poor children is beneficial to them in itself, and there are potentially extra benefits on a range of outcomes. I am a little more sceptical about the idea of it paying for itself, but I do not want to get into a huge argument over that. Let me say where I think it is more plausible and less plausible. I think it is more plausible, if you really have very deep levels of poverty, that bringing people out of those very deep levels of poverty to a more marginal form of poverty is likely to have bigger impacts in general than getting people across a poverty line. I think that there are policies—there are American academics who try and look at every policy and at this kind of question—and those that look like they pay for themselves look a little more like early child development kind of policies, helping people in a holistic way, rather than giving cash. You would have to ask the OBR itself, but I would assume, given how they score things, that they are probably slightly sceptical that these things really do pay for themselves; otherwise, they would be scoring them in that way.

To build on that point, even if you focused only on children living in deep poverty—we are very clear at CPAG that the forthcoming child poverty strategy must be more ambitious than that—we are still talking about 3.1 million children. That number has risen from 2.9 million the previous year. The number of children who are living in deep poverty and experiencing the worst adverse effects of growing up in poverty is sizeable.

Professor Stewart, you talked about why the Treasury does not invest to save. There are challenges with investing to save. The Treasury naturally has a bit of conservatism about putting money in before it can see the results of it. Can you think of any examples of good investment to save? You talked about the electoral cycle—that is obviously part of the challenge.

This is about early years intervention, but there is IFS work showing that Sure Start has reaped long-term benefits, as it was predicted to. We had to wait 20 years to see the full effects of that. That is an example.

That is a good point. It is often a 20-year payback, isn’t?

Yes.

I am delighted to welcome Helen Hayes, the Chair of the Education Committee, as our guest today.

Thank you for having me, Chair. Professor Stewart, I want to drill down into your example about Sure Start evaluation. That shows very clearly, 20 or so years on, that children who lived in close proximity to a Sure Start centre achieved better GCSE results and that there was a direct link between the presence of Sure Start and better exam results. In our report on children’s social care, my Committee drew on similar analysis that highlights the correlation between the decline in Sure Start centres and the number of children entering the care system, which has risen dramatically. I do not think that a straightforward invest-to-save analysis applies, because there is obviously expenditure on those public services and that is a direct cost to the Treasury. We can understand how the Treasury’s accounting processes work. However, I wonder whether a better outcomes-related framework could be brought to bear by drawing on longer-term analyses and learning the lessons. It would have been harder to dismantle Sure Start if that clear, direct link with improved exam results, which are a proxy for economic activity later in life, was known when the decisions to cut were made. I suppose I am asking about the role of longitudinal research. What can we learn from the analyses that have taken place over the last 20 or so years? Are there outcomes-related metrics that we can start to apply to decisions about public expenditure?

I would certainly like to see that. I am slightly sceptical about whether that would have been enough. If we had had those results for Sure Start earlier, would it have saved Sure Start? Sure Start was delegated—the funding was un-ringfenced and allowed to go down to local authorities, and many local authorities that wanted to protect Sure Start were in an impossible situation where they had to meet statutory requirements. Sure Start was known to be popular and there was a sense that it was having an impact. We did not have the evidence. Politics and ideology will always be involved but, that said, I agree that we should think about how to build outcome-related metrics more clearly into invest-to-save models.

It is important to study these things over a very long period of time. The further away we have got from Sure Start, the more we have seen that it was really important. It was quite hard to see that after a relatively short period. It turns out that policymakers in the previous Government moved in a different direction on Sure Start. It is very hard to tell policymakers that we have to wait for a very long period of time to know things. The best way to make policy is to try to design programmes that you think are going to work based on other things and then leave them to bed in for good amounts of time. Looking at very short-run outcomes on its own does not seem to be enough.

There is some work done in Whitehall to make sure that projects have evaluation built in from day one, but you are right, with a seven-year minimum period, impatient politics does not always wait. Sorry, Professor Stewart, do you want to add something?

I was just going to say that I completely agree. We have evidence on Sure Start, so let’s bring back Sure Start. It would save money in the long run; let’s do it and maybe not read too much into the short-term effects on the early years foundation stage profile.

Professor Stewart, one of the justifications for the two-child limit when it was introduced was making sure that parents are more financially responsible and perhaps choose not to have more children. Have we had any evidence since 2017 that that has impacted parents’ decision making?

Decision making around fertility and having children in particular?

Yes.

I have been part of a big project looking over five years at different impacts of the two-child limit, and as part of that we looked at fertility. We only looked at the first three years, but over that period we found negligible effects on fertility. What is nice about our project is that we do quasi-experimental quantitative research, but we have done qualitative work alongside that, and that has shed quite a lot of light on the reasons why and why not. Bear in mind that we have only talked to families who did go on and have a child, but talking to them about their decision making underlines that it is really complicated. People’s lives are complicated and messy. There are a few factors. One is that many parents did not know about the policy. That is something that may have changed over time. However, for many parents, it just was not something they were thinking about, because they were not receiving benefits at the time. Things were good, they had a third child and then the relationship broke down or they became bereaved or a child had a disability and one of them had to leave work in order to look after the child. People are not making decisions about their child with perfect foresight, and that was very clear, certainly in our qualitative sample.

You said that you were speaking to people who had the child. By definition, they are probably more likely to have been less aware of the impacts and not considered them. Do we have any broader evidence of general awareness, and whether people have decided not to have children on the basis of the fact that they have worked out that they would not have the benefits?

I guess that is why we need both the quantitative and the qualitative. The quantitative is telling us that there was not really an effect, and the qualitative is telling us why—why do people go on and have that third child? The third thing I would highlight is that often it is not a choice. People get pregnant by mistake; contraception fails. Also, our sample was about 35 families affected by the two-child limit, and in three cases coercion was involved.

There is an exception for that.

There is an exception, but we found that families do not always want to claim that. It is quite difficult: you have to not be living with the perpetrator and you have to get a social worker or doctor to testify. One mother told us, “I don’t want any risk that my child ever finds out that that was how she was conceived, so I would rather not take the money and not have paperwork somewhere that says that I got that exception.”

Sometimes people do not realise it is coercion; it is just life.

On that point, CPAG took a legal case on the non-consensual conception exemption in the two-child limit. We are in the process of appealing that, but in that process we have had lots of women get in touch with the organisation who are in precisely that situation. They are experiencing lots of practical difficulties, even where they want to access the exemption. As Kitty says, there are lots of women who decide that they are not even going to volunteer that information, but even where people are doing that, lots of practical issues are coming up. It is really important to highlight that you can have an exemption in place, but in terms of how people access it in practice, on something as sensitive as this, it just means that women and children are missing out.

Can I press you on that? It is one of the primary arguments made at the moment; even in politics today, people are saying that parents need to take more responsibility. I admit that this would be a smaller sample, but what about people who have had a third child and then become aware of the cap? Does that affect their decision making about whether to have a fourth? Do we have any data on that, or is it too small a sample?

We were looking at whether you had a third or subsequent—a child who would be affected—and we did not find that there was an impact. There may be more up-to-date research that finds that there is, but I imagine it would be small.

Dr Cribb, I do not know whether you want to come in on this. The data available is from 2017. I am conscious of what you said about how policymakers are impatient. Do you think we have enough data yet to make assessments about the behavioural change caused by this policy?

I do not really have anything to add on that point.

Okay. My next question is about whether the policy makes sense for the country’s finances. Chris Coghlan mentioned one end of the scale, but the other is that we have a declining birth rate in this country and the OBR highlights the UK’s ageing demography as one of our central challenges. Dr Cribb, does it make sense to have what is effectively a disincentive for people to have larger families baked into Government policy?

If it is not pushing down fertility, then alleviating it is unlikely to boost fertility. Irrespective of these problems, it does not seem to be suppressing it. If what you want is a higher fertility rate, it does not seem that this is the key problem with the fertility rate, if you see what I mean. You can only have it one way.

If you accept the evidence we have heard from the panel today, then that makes complete sense.

If it is not having an effect, abolishing it is not a lever to increase the fertility rate.

No. I put the question because it is another argument made, and I just want to make sure that we flesh them all out today. I want to move on to employment, because that is the other end of it. Maybe people would decide to work more if they could. If one parent is not in work, perhaps they might want to start working. If they are working reduced hours, they might want to take on more hours. Do any members of the panel have any evidence that the policy has an impact on employment rates?

We looked at that as part of our bigger project. That was the second key aim—that people would work more, although we should remember that half of the families affected are already working. We looked at quantitative and qualitative data and found no significant effect on the likelihood of working for lone parents or for parents in couples. Another study found a very small effect, but it used very different methodology. Our study found no effect. We did find small positive effects on working hours for coupled mothers who are already working. If you are a coupled mother and you are already working, we found that you would work two or three hours extra per week, and that is consistent with the idea that those families are phasing out of universal credit earlier because of the two-child limit—because their entitlement goes down. We know that universal credit has a negative effect on the work incentives for second earners because of the withdrawal rate. It may be that those families, who are the relatively better-off ones among the group affected by the two-child limit, are finding it more worthwhile and are therefore increasing their hours a little bit. But most families, who are experiencing it just as a big hit to their total income, are not responding by compensating with earnings. People may find that surprising, because it is quite a big hit—it is £3,000 a year. Our qualitative work was really interesting and insightful in terms of what is going on. A small number put a very high priority on being present for their children in the early years. They were saying, “This is the right thing for me to be doing. I’m really committed to this. I think my children need me, even if there’s a financial hit. I’ll move into work once they go to school, but I want to do this now.” More commonly, there were parents who were really keen to work but had faced really significant barriers. That was why they were affected by the two-child limit and why they need the support. Often, they faced significant barriers in terms of childcare; often, there were health issues for parents or children. There were a number of families with a disabled child who said, “I’d love to work, but just right now I can’t. I need to be looking after my child.” We also found families who felt like not having enough money was pushing them further away from the labour market. The up-front costs of childcare, transport, clothes for interviews and so on made it much more difficult, as did, I guess, the time, and the effect on energy of living on a low income. We spoke to one mother who had knitting machines and she had sold them to make ends meet. That is an example of how sometimes not having enough money makes it harder to build a—

Thinking back to my question, I think I have worked out that it has not had much effect on birth rates and it has not had much effect on employment rates. The final stated policy intention is the effect on the welfare bill. Obviously, it will cost money to restore it, but is there any evidence that it has helped to reduce the welfare bill over time, or has it had an impact in different ways?

It has saved about £3.5 billion a year, and rising every year as more and more children are affected, but of course that is having a cost in terms of other public services and there are future costs as well. It is a short-term saving if you look narrowly at the policy, but there are wider costs.

Dr Cribb, there will be a lot of people listening who perhaps are not aware of where the poverty line is currently drawn for a family with two children and for a family with three children. Could we have some IFS-style facts on where the poverty line is today for a couple with two children, after housing?

Goodness, does either of my colleagues have the exact figures?

No.

I have the HBAI figure, which is £550 a week for a couple with two children and around £400 a week for a single parent with two children. Does that sound right?

That sounds right.

That is about £2,200 a month, or a bit more than that, so about £25,000 a year. This is after housing and after tax for a couple with two children?

No, it is not after tax.

Yes, it is. It is net income. These poverty lines are always after tax, after benefits, all income included, adjusted for the family size, and then most of us would then deduct housing costs.

So the IFS would agree that for a couple with two children the number we are talking about is about £2,500 a month.

Yes, those are the Government’s numbers on the Government’s headline statistics, and we use those, along with everyone else.

Can the panel clarify what the amount would be for a couple with three children?

If you can’t tell us now, you can always write to us.

We can write to you, yes.

I just think that it would be helpful for the general population to understand the numbers that we are talking about. The other piece of information that would be helpful—I don’t know if you have this, Dr Cribb—is how clustered a couple with two children would be around that amount of £2,500 after tax and after housing. In other words, what is the distribution of poor families? Are many of them around that income level? Ms Howes, you referred earlier to deep poverty, so for that portion of families who we are talking about who are under the 60th percentile of median income, at the amount that we just said, what does the distribution look like?

I think there are quite a lot of children around that poverty line—a bit above or a bit below. The income distribution does not look dramatically strange on either side; it is not like everyone is clustered just above or below. I do have some numbers. If you define child poverty as below 60% of median income, child poverty is 30%. If you define it as below 50% of median income, it is 20%, so 10% of children have income between that. If you define it as below 40% of median income, it is 10% of children.

A lot of pensioners in my constituency will think that after housing and after tax, having a disposable income of about £2,500 a month would not feel to them like a really difficult level of deep poverty, but you are saying that there is a much poorer group of people. My question to the panel is, would a change in the number of children for whom child tax credits were received alleviate that much deeper poverty? Would that be a meaningful thing? Gordon Brown used to get criticised a certain amount for having “poverty plus £1”, and making the statistics look a lot better by taking people £1 over those cut-offs. What interventions would be most effective for the people—children—in deep poverty in this country?

I will start; there are a couple of things. There was that critique around the 60% of median income and the falls in child poverty levels under the new Labour Government. My memory of that is there were particularly big falls in child poverty on that particular metric, but it was not dramatically different compared with if you went to 65% of median or 55% of median—they were quite broad-based around that part of the income distribution. For people at the very bottom of the income distribution, there are a couple of things to remember. Some people will be there very temporarily, so they will have extremely low incomes but be there temporarily.

Perhaps because they are claiming asylum or are very limited in terms of their incomes?

You get quite a lot of people who are self-employed or starting up a business, so they have very low earnings for a while, but they are not actually particularly poor. I am just clarifying, before I make a further point, that there are some people at the very bottom of the income distribution who do not look that poor. Let us put them to one side for the moment. Once you go to the very bottom of the income distribution, you are even more disproportionately focusing on workless households.

Persistent, deep poverty.

Yes, and when we think about in-work benefits, working more hours or a higher minimum wage—all these things that may play a role for people around the poverty line—that is quite different when you are focusing on the very poorest who, as you say, are more persistently out of work. It is fair to say that although the welfare system is not the only lever that the Government can or should be thinking about in this system, social security is going to play a particularly important role for this kind of persistently poor, out of work group.

Ms Howes, do you want to add anything on the point of deep poverty?

To make the obvious point, for the families who are in the deepest poverty, that correlates with higher levels of need and therefore higher costs—for example, larger families or families where disability might be a factor. I think what you said about migrants is a huge factor—people who have no access to the social security system. JRF has done some quite interesting work looking at the correlation between those families who are in very deep poverty and the high numbers of those who would be migrants. In terms of interventions that target families in deep poverty, we have talked a bit about the two-child limit. It is worth highlighting that removing the two-child limit would be the most cost-effective way of lifting families out of poverty—350,000—but it would also mean that 700,000 children would be living in less deep poverty, so that would be very effective in terms of deep poverty. The other policy is the benefit cap. That affects a much smaller group of families, but they are living very far below the poverty line. If you scrap the benefit cap, those families would still be living in poverty, but closer to the line. It is really important to remember that policies targeting families living in deep poverty need to be an important part of the picture and the forthcoming strategy.

Can I give a number on that? We think that if you remove the two-child limit, it would reduce deep child poverty by 110,000. If you remove the benefit cap and the two-child limit, we think that would go to 200,000, so it would almost double the impact on deep poverty.

Where are you drawing the deep line?

What we have done is below 40%.

Ours would be below 50%, so that is why the numbers are higher.

So you will follow up with some facts about the income level required to go from about £2,500 a month after tax and housing costs for a couple with three children?

We can write to you on that.

Thank you.

Just to follow up on that, I can understand that a pensioner might look at that and think, “That’s a lot of money,” but what we are doing is equivalising. We are trying to adjust for the household needs, and so we are working out what income that family would need, because they have the higher levels.

I understand; I was just asking what the difference was.

In your answers to my colleague Bobby Dean, you suggested that, in essence, the qualitative and quantitative data demonstrates that the two-child policy has very little behavioural effect. We also discussed the challenge of short and long-term changes. There is evidence to suggest that, if you invest a lot in the short term, you could have a change over time, but the reality is, whether rightly or wrongly, the way that our politics and cycles are set up means that every fiscal event has a lot of pressures. How do you see the Chancellor’s choices of direct money transfers to alleviate child poverty versus longer-term choices about inducing behavioural changes—doing things that grow capacity for people to help themselves? What is the most effective use of very limited resources in these fiscal circumstances to actually alleviate poverty?

I think it is very clear that it is much more cost-effective to make the transfers to the families with children. That is partly because, if the alternative is to invest in employment support, which I would also like to see, all the estimates suggest that that is really expensive. We should be doing it, but given what Sophie said earlier, a lot of parents have moved into work over the last decade—it has been quite a success story. The parents who are not working are often facing significant barriers to work. You are talking about quite extensive personalised support, so there is a lot of money needed there. All the simulations done by the IFS and the Resolution Foundation show that, even if you get parents into work, actually, in terms of reducing child poverty, it gives you much less bang for your buck than investing in the social security system. I understand that some may think, “There are people sitting at home doing nothing, and we are just giving them money. Isn’t it better in the long run to encourage them to go out and look for work?” I think that is a false representation of what we see. A lot of families who are in work—who are working the hours that they can, given their caring responsibilities—are really struggling, and that money would make a difference and help them to invest in their children. For other families, working is difficult in the short term—because of, for example, a disabled child—and supporting them through that period, which is what a social security system is designed to do, may even make it easier for them to move into work in the long run.

So essentially this is a prerequisite for progress and social mobility, because you will not get that far unless you remove the poverty in the short term. That is what you are saying.

Absolutely. That is a much nicer, more succinct way of saying it.

I am very rarely succinct, as Meg points out. Dr Cribb, would you like to say anything complementary or additional to that?

Clearly we are in a difficult fiscal situation, but the amount of money we are talking about, in terms of abolishing the two-child limit, is not so dramatic that it would not be the kind of policy that even fiscally constrained Chancellors have chosen to do, even at difficult Budgets. If this was £20 billion, or even £10 billion or £15 billion, it would be a lot harder to bring together a package that meets the Government’s objectives. If a Government’s No. 1 priority was to abolish the two-child limit, then given the scale of this—about £3 billion—the fiscal position of the country does not preclude it. That is what I am saying.

Recent Governments have used the household support fund as a sort of top-up mechanism. How does that equate, given it allows more discretion at the local authority level? Is the argument that it is just not of a quantum that is big enough to move the dial?

My knowledge of the household support fund is not enough for me to give you a good answer on that. I can write to you, if you want, or other colleagues can come in, but I do not want to say things I do not know.

No, that’s fine. Ms Howes, you were keen to come in.

On your point about the two-child limit versus other measures, to reiterate, we would see it very much as a foundation on which to build. If you play out the scenario where you do not scrap the two-child limit and the benefit cap, and you invest in other things—maybe non-social security related things—the reality is that that will act as a drag on those interventions. Those interventions will not be as effective. We would very much see it as a foundational piece to this forthcoming strategy. On the household support fund, I would have to remind myself about the scale of investment—I think it is about £1 billion.

Recurring each year—rolled over.

Yes, each year, to local authorities. It is a discretionary payment. We know from our work with families that there is lots of trouble accessing it. People are not aware of it. It is just not able to provide support in the systemic way that we know children and families need. While it is an important part of the social security system, and it should be there for crisis situations, the idea that it could plug the gaps in an inadequate social security system is flawed.

On the basis of your research, what would a truly ambitious approach to tackling child poverty look like? It seems to me that we are talking about how to get over a certain baseline. Given the way that poverty is measured, when we talk about lifting children out of poverty we are actually talking about lifting them from something that is very definitely poverty into something that probably still looks quite a lot like it. We are not talking about raising the level of household income to a particularly comfortable standard, and we are not necessarily talking about all the other improvements in support and access to services that might make a substantial difference. On the basis of your research, what does ambitious look like in this space?

Just to clarify, when you say that it would not be transformative for families, do you mean just getting rid of the two-child limit, or getting rid of poverty but—

I am not saying that relieving the two-child limit would not be transformative—I think it would be—but I think we would still essentially be seeing lots of children living in households with very low levels of income, albeit above a technical definition of poverty. What I am interested in is this. If we accept the baseline, where do we go next, above and beyond, if we are aiming towards a country where every child is really thriving?

Scrapping the two-child limit and the benefit cap, which is a really important part of the story, should be just the very first step. Then there is much more we need to do to build a more comprehensive social security system, and I think that would start, first of all, with looking at local housing allowance. We are targeting after-housing-costs poverty, but it is really important to think about housing costs, because that is a really important part of the story. We also need to look at the wider uprating of benefits over time. Just as the triple lock has helped the state pension to really get to where it is a reasonable level of income, we need to do the same for child benefits. We should think about where the floor should be and design our social security system so that we do not need to have a big argument about it every year, but instead build that in, so that we are heading in that direction. Universal child benefit would also be a part of that picture, if we want every child to thrive, recognising that all families need support for children at some stages in their lives. That is social security, and that would be about the income. If we saw all those things in terms of family income, that would be hugely different from where we are today, but then we would also want to be looking at childcare, housing, early years and education as the four planks, recognising that it is difficult at the moment—we have just been talking about fiscal constraints.

Dr Cribb, the IFS produced a very good table in your October 2024 report, “Child poverty: trends and policy options”, which indicates that once the two-child limit is fully rolled out, for every category of child, whatever their ethnic background or housing background, or whether it is lone parents or couples in work or out of work, the share affected goes up. We have talked a bit about other long-term options, around housing and work and so on, which Mr Dean highlighted, and you have been clear in your position at the Child Poverty Action Group, Ms Howes, but I wanted to give you an opportunity to talk about anything else that could be done to keep that figure down. You have been very clear that the two-child cap is the main thing, but can anything else be done in the short term? Given that the Chancellor is going to face choices at the Budget and a lot of pressure on this issue, for all the reasons we understand, is there anything else that could be done at the Budget that would make a significant difference, picking up on what Mr Glen was asking?

I think the two-child limit needs to go, and it needs to go because that is going to stop child poverty rising. It is probably worth being really explicit about that. It would lift hundreds of thousands of children out of poverty, but CPAG estimates that in the alternative scenario, where that does not happen, child poverty will reach 4.7 million. If we want to look at equivalent reductions in child poverty without scrapping the two-child limit, it is really difficult to do that. We have looked at alternatives within social security spending, and you could get an equivalent number of children out of poverty if you put money into the child element of universal credit, but that would cost 50% more than the two-child limit. You could increase the standard allowance, but that would cost £8 billion to get 350,000 children out of poverty. The alternatives in the social security space are more expensive, so doing this is definitely the most cost-effective option, in terms of the number of children lifted out of poverty per pound spent. If you are talking about wider measures around housing, childcare or those things, there are no equivalent things that you could do in the short term. Those things should be a really important part of the strategy, as Kitty has said, but they are things that will reap results in 10, 20 years. That is just the reality. So we really need the short-term bit—I am reluctant to say that social security is a short-term thing, because we really want people to view it as an investment in children, and it is something that will always be there—but this is the thing that is going to make the critical difference now.

Do any of the other panellists have anything to add to that? Any quick fixes for the Chancellor?

I would definitely highlight the benefit cap as absolutely essential. The two-child limit without the benefit cap will leave many families in very deep poverty. Local housing allowance would be the next thing on the list.

All the quick fixes are in social security; employment and other things are going to take too long to deliver in this Parliament.

Yes, they are expensive and longer term.

Just to reiterate what has already been said, the majority of families that are affected by the two-child limit are working—59% of them are working. It is 450,000 families that are affected, and the most recent set of data published by DWP provides a breakdown of conditionality status: it tells us what they are required to do in terms of working more. What we see from that is that of those 450,000 families, only 160,000 are expected to be looking for work or looking for more work. The other 290,000 are already working enough to escape the conditionality regime, or they have got very good reasons for not working. They may have a very young child at home, they may have a disability, or they may have a long-term health condition. That is an important part of the picture to understand. We have already talked about black minority ethnic families. Kitty has said that they are three times more likely to be affected by the two-child limit—so they are disproportionately affected by the policy. It is worth flagging the regional variations. If you look at the different parts of the country where child poverty rates are the highest, London, the north-west and the west midlands are the three regions with the highest child poverty rates, and there are constituencies where over 50% of children are in poverty in some of those areas. That is compared to the average, which is around 31%. There is a strong correlation between those child poverty rates and a high prevalence of families affected by the two-child limit—so we can see that it is a strong driver. You will be aware that there is no exemption in the two-child limit for disability, so it is affecting families that have very high barriers to work in an indiscriminate way.

Yes, definitely. There are some questions about the rationale for the policy. The policy was born out of this conceptualisation of benefit claimants as a static group of people who are not working and are making a decision about a third child while in that situation. We know that it is a much more fluctuating population. People are claiming benefits for a whole range of reasons. When we look specifically at families affected by the two-child limit, the vast majority are either in work or have a very good reason for not working.

I was going to ask about the agglomeration effect of London and the north-west, but you have actually answered it. I will quickly ask about a benefit that used to be called educational maintenance allowance. You mentioned older children, and that did have an effect on larger families where there were two or three teenagers at the same time. It meant that there was a greater retention in college or sixth-form, and a bridge into university. Looking at university retention rates is another good example, in that it is pretty expensive to go to university, but it has a long-term value on confidence and all those other things that are hard to quantify but which probably lead to improved outcomes financially in the long run. Has any research been done on that 15 to 19 group? Educational maintenance allowance used to help a lot of people in that grouping. Of course, it was quite specific to certain parts of the country, but in my case, 75% of children in the borough where I was leader were affected by that one change. That is three quarters of families just being whacked at the same time as everything else, with LHA and all the rest of it. Has any research been done on the impact of taking that away from this particular group of people, or how the impact on their life chances can accrue?

CPAG has not done any research on the that. My recollection—I am hoping Kitty might help out—is that Jane Millar, a colleague of ours, has done some work on the educational maintenance allowance, and she is an academic based at the University of Bath. I do not know whether that is triggering any memories for you, Kitty.

I know that it was evaluated, but that was before it was a scrapped.

We need to look into that then.

That is an example of a policy where we did know that it was working. We had the evidence at the time, but it was still scrapped. I guess that plays into Debbie Abrahams’s question about NEETs. We can look into it—there is some interesting research to do there.

If you can point us in the right direction, it is always helpful for us. Thank you.

We can follow up.

Just to start, I have a very quick yes/no question to summarise a lot of what we just discussed. Do you believe that lifting the two-child limit is the most effective way of reducing child poverty with the amount of money that it would take? I am thinking about its cost-effectiveness.

Yes.

Yes.

Yes.

There has been a lot of speculation about alternatives to scrapping the limit, including different ways of tapering or adjusting the limit, which would save small amounts of money in different ways, but they might reduce the impact. Do you want to speak to the viability of those alternatives and whether they are cost effective, compared with a complete scrapping of the cap?

These are the five-child or three-child limits, for example.

For example, yes.

Lots of things have been talked about, such as a three-child limit, or scrapping it if the person is in work, or the scrapping of particular ages. Broadly—this sounds a bit crass—if it costs about half as much as scrapping the whole thing, it has about half as much of an effect on poverty. You can draw a line; if it cost three quarters as much as scrapping the limit, it will have three quarters of the effect on poverty. Who exactly it affects will come down to exactly how you do this. If you introduced a three-child limit, it would clearly affect people. It would loosen this for everyone, but the people with four kids and more would still be capped. You could design it in lots of different ways, and we could talk about that forever, but I think the big point is that all of these different things are halfway houses between either having it or not, on both cost and effect on poverty.

If the objective is to make sure that child poverty is lower at the end of this Parliament than the beginning, do any of those halfway houses get us there, Dr Cribb?

I can’t exactly say, but you are certainly less likely to get there if you only go halfway.

Does anyone else want to come in?

One thing that has been playing out in some of the debates about compromised positions on the two-child limit is the idea that there is some kind of economy of scale with raising children. We published some research on this last week. Every year, the Child Poverty Action Group publishes a report called “The Cost of a Child”, which basically looks at the cost of raising a child to 18, in partnership with Loughborough University. That shows that there is a slight difference between a first and second child, but once you go from second to third, fourth and subsequent children, there is no notable difference. Savings that can be made on things like reusing items are very marginal. What really costs with children is things like childcare—it is those really bulky costs. There are really no differences, so I think it is important for the Committee to understand that.

We have just discussed the benefit cap and the impact that has on the poorest families. Dr Cribb, in the IFS evidence that we have received, there is a statement saying, “But the benefit cap would wipe out the gains”—of reversing the two-child limit—“for some children in the very poorest families.” That is talking about the deepest poverty. Professor Stewart, you just mentioned the different ways in which the pensions and benefits systems work, and the way in which, for example, pensions are pegged to a triple lock, rising with inflation, earnings and so on. For the benefit of those not aware of how our very complex benefits systems work, could you explain how the benefit cap and the different kinds of child-related benefits we are discussing move over time?

The thing about the benefit cap is that it places a total cap on the amount that a family can receive, if no one in the household is working or is working fewer than 16 hours. That includes housing costs. It is predominantly affecting families who have high housing costs, and it means that the social security system does not pay anything like the full housing costs. Those families are the ones living in the deepest poverty in our country. We have done research on that group, and the level of destitution that some families are living in is really shocking. A family of five is living on £65 a week after housing costs, because the benefits are not covering that cost. The point is that, if we lift the two-child limit, that family is already capped, so they will see nothing as a result. You are taking out those families—100,000 families at the very bottom. Lifting the two-child limit would affect a lot of people across the lower part of the distribution, but you are taking out the very bottom.

And the cap itself does not move automatically over time?

No, interestingly. It was introduced at a certain level—£26,000—and then it was actually lowered. I think it has only been uprated once with inflation since then. Every year, it bites more and more, and it is pushing people into more and more hardship.

So over time, the number of families affected by it will simply increase, unless it is changed in itself?

Yes—the number of families and the depth of their poverty.

Finally, what are your range of costings for removing both the two-child limit and the benefit cap?

These are a little out of date because of inflation, but, back last summer, we thought that removing the two-child limit would cost £2.5 billion. That is probably closer to £3 billion, once you look towards the end of the Parliament. We thought that adding on the benefit cap would cost an additional £900 million—so maybe a further £1 billion in addition to the two-child limit now.

Would you agree with Professor Stewart that we should see a more fundamental reform of the benefit system, so that we do not need to have this uprating conversation every few Treasury Committees?

Freezing thresholds, whether it be in the tax system or in the benefit system, is usually not a very good long-run policy.

What would you like to see instead?

I do not have personal recommendations for what I would like to see.

It is a broad question.

At a minimum, you would expect these things, in general, at least to rise in line with prices. Some Governments could then either index it to something more generous than prices, such as earnings, or raise it in line with prices, with concrete steps to increase it at particular points, depending on their political persuasion.

I think we have probably had an answer to this—certainly very clearly from Sophie Howes. The Chancellor has choices—we have touched on that a bit, but perhaps not as much as we could have done in this hearing. You could remove the two-child benefit cap, but we have talked a lot about children in really deep poverty. If there were a difficult choice, what would be better: to take the children out of the deepest poverty, or to lift the two-child cap and affect more people? It is a hard choice, but these are the realities of what the Chancellor will face.

It is not an either/or. The two-child limit would mean that 700,000 children would be living in less deep poverty, so it would have a substantial impact on deep poverty.

Professor Stewart?

Are you essentially saying, “Benefit cap or two-child limit?” if we had to choose?

You could cast it that way.

I would really hate to see it as an either/or. Both policies are doing a lot of damage, but I think the benefit cap is doing very severe damage to children, and that would be my No. 1 priority.

Dr Cribb?

I don’t have recommendations here. There are differences in the choices. There are differences in who is affected; there are also differences in scale. Doing one costs about £1 billion and doing the other costs about £3 billion. Do you want to spend £3 billion more or £1 billion more?

These are the very difficult decisions that the Chancellor has in front of her.

One thing to note is that if you lift the benefit cap, you really will not affect your child poverty headcount, which I know the Government want to do something about.

As I say, the Chancellor has challenging economic and political decisions to make ahead of the Budget. We hope that this session and our witnesses have shone a very important light on the issue of child poverty, which is a very big issue of debate. Some of us are living in hope that something may happen at the Budget, but that is a matter for the Chancellor. I should stress that even on this Committee we have no inside track. I thank our witnesses very much indeed for their time. We have had a really useful discussion and analysis of the available evidence on the impact of child poverty and low income on life chances and outcomes, which is very important to have laid out. We have had a discussion about the role of employment support—however important that is, it cannot act as quickly as some of the other things we have been discussing. We have had an examination of policies to address child poverty, including the potential for employment support to reduce that poverty, and we have discussed the cost of policies to address child poverty and the options of the Chancellor. Thank you to our witnesses, Dr Jonathan Cribb, Professor Kitty Stewart and Sophie Howes. Hansard has been reporting the session, and the uncorrected transcript will be available on our website in the next couple of days. Thank you to them and to our colleagues at Bow Tie for the broadcasting. Our witnesses will be writing to us on a couple of points, so thank you for that. Thank you very much indeed for your time.