 Welcome back everybody for the last session. So first off, we've got Susan Harkness. Susan is a professor of social policy at the University of Bristol, visiting professor at the Center for Analysis and Social Exclusion at LSE. And the research lead for labor markets and institutions at the ESRC Research Center for Microsocial Change at the University of Essex. Her research examines how gender and family structure relate to labor market opportunities, income, inequality and poverty. So over to you Susan. Thank you. Okay, hopefully you can see my screen. Yeah, we can. Okay, great. Okay, thanks very much for the introduction. So as you said, I'm Susan Harkness. I'm at the University of Bristol and today I just want to talk about gender equality within the household. So I think this is something that is perhaps a little bit under-researched and has potentially important implications for how we think about poverty and so on. Let me just quickly go to the motivation. Actually, before I sort of get too far, let me also just say that this is very, very much work in progress and the results are very preliminary. So certainly not for a citation at the moment and we're still very much working on thinking, well, thinking about the data and also the modeling. Okay, so in terms of motivation, we know that traditional models of household protection, of course, assume that men specialise in work and women special in Eisenhower protection, particularly after children are born and that resources are shared within the household. Now, of course, as more and more women have entered into the labor market and in particular, mothers have increased their labor force participation, we know that this is going to have affected both the composition of family income, but also potentially the way in which families organise their finance. So that's essentially what I want to look at today is whether women's increased participation or actually more specifically women's increased earnings is affecting the ways in which families manage their finances. And there's been some past work on this. So in particular, the work of Manny Kan and Heather Laurie looked at, for example, changes in savings and investment holdings of men and women in couples. And recently, Jan Hu has also looked at cohort changes in patterns of money management. Okay, now, why do I think this is important? Well, one of the reasons it's potentially very important is because whilst we've had the introduction of independent taxation in 1990 where women are treated separately in terms of their assessments of income, as recent changes over recent decades have meant that we are seeing for those on low and increasingly middle incomes are increasingly being drawn into means testing, particularly if they have children. And this means that people's access to a income is really very dependent on what's happening within the household. So it's potentially important implications for what happens within the households and money management. Okay, so the specific question I want to think about is whether the sources of household income affect patterns of sharing. So my research questions are essentially two. The first is, have the ways in which couples manage their finances changed over recent decades and does the source of money affect the ways in which finances are organized? And then I also want to think a little bit about whether this matters, does it matter how household finances are organized? And I'm going to think about this in two ways. I'm going to think about, first of all, how it impacts individuals' perceptions of financial stress. And secondly, how it impacts their psychological wellbeing. Of course, there are many other ways you could think about this, but they tend to be including measures of deprivation, but unfortunately in the data that we have, that's not something we're able to do. Okay, just very briefly in terms of prior research, we know that there's been considerable attention to gender pay gaps and the dig motherhood pay gaps, but there's been far less attention to what those changes in gender and motherhood essentially wage gaps, which are typically looked at, has for how individuals do within the household. So in particular, what could want to be concerned about is financial inequalities within families. And we know that financial inequality within couples matters. So there's work that goes back to the 1980s and 1990s, particularly the work of Jan Powell, which I think was quite influential in how we think about female poverty and suggesting that the distribution of resources within households matters. And more recently, the work of Elina Karakiannaki and Tanya Bokat has looked at deprivation with own households in Europe. And they essentially find that if we looked at deprivation and measured at an individual level, we would have considerably higher estimates of the numbers of people who are deprived. So again, suggesting that household resources are not always equally shared. And then there's also increasing interest in this question. I think in the US, we know that financial inequality has wider sets of outcomes. It's linked to poorer relationship quality, perhaps unsurprisingly, it's also linked to divorce when, for example, savings and investments are held in separate names, but there's also links, wider links, I think, to indicators of child wellbeing and what is spent on children. Okay, what else do we know about financial inequality? Well, we know that Jan Powell and Karl Vogler argued that it was financial inequality varied across household income. And she actually argued that financial inequality was greatest amongst those with lower incomes. And she also found that having access to an independent income did not necessarily improve equality within the household. So of course, this was data from some time ago and female employment has changed considerably since then. So what I want to think about is how these findings sort of resonate today. And the other area I want to look at is how changing family forms matter. So in particular, we've seen that a rise in cohabitation, so move away from marriage, and an increasing number of sort of complex families, so by which I mean families where there are stepchildren, there may be children belonging to both parents or biological parents, but there are also increasingly families where there are stepchildren present. Okay, and evidence from the US suggests that in these families, there may be less sharing. So I want to also have a look and see whether that's the case in the UK. So data, the data comes from understanding society. And I'm going to look at couples, only those with children. Most I'm going to focus on the results for women to look at the sharing outcomes. I'm also, in some of the final results, I'll also look at how sharing effects outcomes for men. And just to say that sharing information that we have is available in 2012 and 2016. That's the data I'm going to focus on. I'll tell you a little bit more about it in a moment. And we have reasonably large sample sizes, so it's around seven and a half thousand in those two years for women, and a similar number for men. Okay, so when we started off thinking about this, we wanted to look at two indicators of financial arrangements. One was money management, and that's what I'm going to look at today. We also wanted to think a bit about savings and debt and who holds savings and debts within couples. But we haven't actually managed to really get to grips with separating the data in understanding society. And we may find that in fact, we will be looking at other data sources. Okay, money management. So a variable on money management is available. There's a question asked on money management in the early part of the BHPS, and then again in 2006, 12 and 16. Now, the years 2006, 12 and 16 have a consistent measure. There's a slight change in the variable beforehand, so I'm not going to use that information. I've used this question on who manages the Hustle finances to code four separate variables, which are really very, very similar to those that were originally collected. Hustles or individuals are asked about how closely their perception of money management relates to these various forms of money management, and the options are there's complete pooling. So we've got joint financial management. There is a female control where the woman manages the money and the man is giving spending money, or this is also merged with data on whether they have a housekeeping allowance, which obviously now seems quite old fashioned, but that's also an option that was allowed. And then we have male control, so the men manage money and women get spending money. And then we have a category which I've got to call independent or a separate finances, where we have either complete separation of finances, which is in fact very, fairly rare, or we have part pooling and the literature around this suggests that when money is partly pulled, this is typically reflects independent sort of decision making about money and spending in the Hustle. So we have this category of independence or which we can also think of as a largely separate decision making over finances. Okay, in terms of models, I'm just going to run multi-loan me allergic models against the base where income is pulled, where we look at the sort of types of high sold financial arrangements. So that's the first type of model I'm going to run. And then the second type of model I will run is to think about how these financial arrangements affect individuals perceptions of financial stress and psychological wellbeing. I'm going to have to go very quickly to just to get to the results. So in terms of variables, we've got a control for family income. These are the ones I'm going to focus on. The share of income from own earnings and the share of income from benefits. And then I'm going to also look at the results for cohabitation and whether the individuals have step children, in fact. I'm going to control for a range of other factors, so the age of the head and spouse, number and age of children, relationship satisfaction, duration and ownership. And then finally, I'm going to run these models on these outcomes, where I'm going to think about financial strain. I'm going to focus on the outcomes of whether individuals are finding it quite difficult or very difficult to get by compared to living comfortably or doing all right. And then I'm going to finally look at an indicator for depression based on the GHQ score. So the first thing to note, and this is changes over since 2006, is that money management amongst couples has shown some small but pretty gradual change. Most couples pool their money and in fact, if anything, pooling has slightly grown over the decade. The share where either women manage the household finances or the man manages the household finances is slightly declined. But what we also see is a small increase in the separation of household finances. Okay. And we can also see that there's some gradients across income desks across the distribution with more separation of finances amongst higher income families and women more likely to control the household budget in lower income families. Okay. Now the results from the model, so these are just really just picking out the coefficients that I want to focus on. So the first thing to note is that if we think about predicting money management, we can see that, if we can see that in particular, if we focus on the separation of household finances, we can see that some things matter. So having a higher income leads to a greater separation of finances. Women having a higher earning share also leads to a greater separation of finances and benefits, benefit received, the greater the share of income from benefits reduces the probability of having separate finances. The other result I want to focus on is this female control and female control is also more likely where women have a higher share of the household income but also where a higher share of that income comes from benefits. And indeed, the female share of income is really strongly related to men not controlling household finances. Okay. In terms of family structure, what matters? Well, the results I just wanted to pick, sorry, pick out here are again related to the separation of finances. And here we can see that when we have cohabitation relative to marriage, so cohabiting parents are much more likely to have more likely to have separate finances that those that are married and step children, the presence of step children, even in married couples is also really strongly linked to finances being run separately. And this again is after controlling for our whole host of features including perceptions of relationship quality, for example. What does this matter? So if we think about how this affects individuals outcomes one of the things we looked at was whether it affects perceptions of financial stress. And what we find here is that where couples keep the finances separate, women are more likely to report difficulty managing their finances. Whereas for men, there is a sort of negative coefficient. So if anything, the direction of separating finances is perhaps associated with men having less, being less likely to say they're struggling financially. In terms of mental health, however, we find that separate finances is associated with sort of positive effect done worse than in mental health. So it's associated with a greater risk of depression, more significant for women than for men. Okay, so just to conclude, what we find is that women have entered the labor markets, not finding, and we show that money management has actually over the last decade has changed slightly, but not very much. Women in low income families and those who receive a higher share of income from benefits we find are more likely to manage the family finances. And in some ways this seems to, certainly relative to separate finances seem to protect women from greater financial stress. But it should be noted that other studies suggest that this doesn't necessarily mean it protects them from deprivation. We know that women, for example, make choices and that those choices may also mean that they're more likely to be deprived of certain items. Women's increased earnings are associated with greater separation of family finances. And again, this resonates with Jen Pannell's earlier study where essentially you might find that whilst higher earnings and indeed higher income are associated with greater separation of finances. And that may mean that those in higher income households are also at risk of being deprived even if their household income is above the sort of lower level, poverty levels. And then finally, those who cohabit or who have stepchildren are more likely to keep their finances separate. And I think this is potentially important implications, not least because the cohabities and stepchildren families are more likely to be represented in the lower part of the income distribution. Okay, and I will finish that. Thank you. Thanks, Susan. Thank you. Okay. Up. Next. Anna Nicarico. Anna, are you there? Oh, hi, Anna. Okay, so I'll just introduce you. So Anna is a PhD student at the University of Manchester and part of the Doctoral Training Data Analytics Society. Sorry, that doesn't make sense. With a background of economics and data science, her PhD focuses on studying difference in terms of poverty and material deprivation in the UK between families raising children with disabilities and families raising children without disabilities. So if you're ready to start, Anna, that would be great. Okay, yeah. So just as it was mentioned, I'm Anna. I'm a PhD student at Manchester University. And just the topic of the presentation is to look and make a comparison between families with children in the UK, between those who have children with disabilities and those who don't. And I'm gonna look at income poverty, material deprivation and also paid care using data from 2019. So just an overview. I'm just gonna go quickly to some previous research. Then I'm gonna present the research aim of this analysis, then the data sources, some results and just a brief conclusion and then answer some questions. So in terms of previous research, well, obviously there's a lot of talk about poverty, but in 2018-19, which is the same year that I used for my analysis, the Social Metric Commission estimated that there were around 14.4 million people in poverty in the UK. And it's also at that time, 33% of children were in poverty. But what's interesting to look at is how this is split. And we can see in the bottom half of the table that actually half of the people, so 7.2 million were people in families that had someone that had a disability. So either it was an adult that had a disability in the family or a child or both an adult and a child. And I guess there are various reasons why that can be and it is known that there is a connection between poverty and disability. But one of the reasons could be to do with unpaid care, which is very hard to find in a way and it's also I think harder to assess and capture in a survey or at least that was my, that's what I found from my research. But I guess generally unpaid care is any work that you take time and energy and you take care of someone and you're not getting paid for it. And while it's very beneficial for the society at large, it can have very negative impacts on the person who's providing care. So for example, it can reduce the amount of time you have to be, to do work, to be employed. So then it has impact on your employment and your income. For example, 22% of informal carers were living below the average housing cost income in 2013-14. There's also the aspect of mental health, not just financial well-being. And in extreme cases, it can lead to social exclusion. And there's also, we can see that informal care in 2016 was estimated to be around 351.7 billion and this is money that should have been paid and the caregivers should have received some money and help. But this is something that's the society in a way to them pay for them. So it was free work and it's important to take this into consideration. Maybe consider in the future what we can do about this. And this is because I think at some point in our lives, everyone has to provide some form of unpaid care and there's various reasons why you would have to do that. And one of those reasons could be disability, which again in 2017 or 18, if you're looking at disabilities for children, families with a disabled child had a poverty rate of 26% compared to 19% of families in which no disability was present. And this could be because someone had to provide unpaid care. So they weren't working or it has to do with the cost of disability, which for example, it's estimated to be around 581 pounds per month, but it can even go higher up. So in a report in their surveys, 24% of families, the disability costs went over 1000 pounds per month and these are costs from which any disability benefits were deducted. And I guess this is to consider that some people might not be considered poor because they have the same income level, but obviously their disposal income is different from a family who doesn't have children with disabilities. And I guess these were just some examples to look at the connection between poverty and unpaid care and poverty and disability, which has been researched before, but this also leads to the research aim, which was to try and combine the three of them together. So I wanted to explore the three-part relationship between poverty, unpaid care, and children with disabilities and their families and look at various socio-demographic variables and the effect they have on poverty. So to do that, I used the Family Resources Survey and it was around 551 families with dependent children. And this is, when I talk about families, it's not the larger family, it's the very strict family in that it's just the parents and the child. So even if, I don't know, maybe you live, the grandparents live in the same household, they are not considered, they are not taken into consideration in this analysis. So the data was split in various groups than the ones of interest were disability and unpaid care. Disability is defined according to the Quality Act and it's just if you have any illness that's expected to last 12 months or more and limits your ability to carry out day-to-day activities. In terms of unpaid care, there's no fixed definition and it can be for many hours or only a few hours a week and included examples such as going shopping for someone or helping with paperwork or gardening, but this is, if you're looking at children in a way, you are already doing these things for them, but in the FRS, it's also mentioned that unpaid care for people who have some form of disability, but it's debatable on how well these measures unpaid care for children and what's the difference between the normal childcare and unpaid care just because the child has extra needs. In terms of measures and outcomes of the analysis, there were two of them. So it is income and child material deprivation. Income is the weekly net income. It uses the equivalent scale and the threshold was used for 60% of median income and the material deprivation was discussed before. It's just 21 questions about kids and services and it's just a score calculated and if the score is 25 or above, then the children and their families are considered materially deprived and these were treated as binary variables. So logistic regression was used to try to predict after housing cost income and material deprivation and then the data, we had disability versus no disability for children in the family and unpaid care versus no unpaid care and these two were treated as both binary regressors in one analysis and then the data sets were split into these groups and then the social demographics. Effect on poverty measures were considered and compared across groups. So in terms of results, in the first part, we were just looking at whether belonging to either group, so disability or no disability has an effect on poverty and this is looking all the families with dependent children and we can see here that in terms of after housing cost poverty, which is a yes or no classification, we can see that actually having children with no disabilities seems to be increasing the odds of poverty by 2.17 times, which is quite surprising as we were expecting it to be the other way around and in the second half of the table, we are interested in these results and this is just family supporting children with disabilities and we're interested in the effect of unpaid care provision and in this case, we can see that again, surprisingly, providing unpaid care has lower odds of being after housing cost poverty, which you would expect again to be the other way around that, but yeah. Then in the second part, we only looked at the difference between effects of various social demographics such as the type of job you have or where, whether you rent or what's the parent's education and the family education level and in this case, we look at families who support children with disabilities and those with disabilities, so the data was split into two groups. In here, just a few points. I guess we can see that higher education has lower odds associated with material deprivation for those raising children with disabilities compared to the first group and however, being of any other ethnicity that might increases the chances of material deprivation by 3.2 times for those supporting children with disabilities. So there's not a much difference between all the other social demographic variables available and finally, the last part was to try and look at those who only have children with disabilities and they are split between those who provide unpaid care and those who don't provide and in this case, no major differences were observed between the groups so it doesn't matter, the results were not significant in terms of what education you have or ethnicity, the, especially when we're talking about the after-housing cost poverty. The only difference would be between the type of renting that you have and that for those providing unpaid care, private renting increases the odds by 7.32 times compared to only 4.6 times for those not providing unpaid care. So just these two and I guess overall, just we found that social demographic factors have an effect on children's material deprivation when groups are treated independently but in the other hand, they only have the groups so whether you have a child with disability or not or whether you provide unpaid care or not has an impact on after-housing cost poverty when the groups are treated as direct effects and this I think is just an initial exploration and there were some interesting results and I think future research needs to try to untangle some of the points and see how we can maybe better understand unpaid care for children with disabilities and how can this be measured and assessed. That's all. Thank you, Anna. All right, next up we've got Lynn Tian. Lynn is a researcher at the Centre for Personal Financial Well-Being at Aston University. Her research focuses on household finance, financial inclusion, labour market outcomes and some topics on social housing and subjective wellbeing and today she's going to talk to us about what happened to the Child Trust Fund. Welcome, Lynn. Hello, Carla. Hello, everyone. Okay, so the paper I'm going to present today is about the Child Trust Fund and Children's Service in the UK. This is a paper called, so it was Professor Steven McKay from University of Lincoln and Professor Andy Leimer. Both of us work for Centre for Personal Financial Well-Being, which was recently launched at Aston University. And Steve is also here today and will join me in the Q&A. So for the presentation, I'd like to start from an introduction on the Child Trust Fund and then we will focus on a set of research questions and the univariate and multivariate analysis we did to answer those questions. And finally, we will draw the conclusions and the policy implications. So the Child Trust Fund originated from long existing ideas about lump sums given at adulthood. This idea was brought to light in 20th century and was developed into the asset-based welfare theory. So the basic idea is to redistribute the productive assets to narrow down the gap between the rich and the poor and to eliminate poverty in a preventive manner. So Child Trust Fund is one of the first asset-based welfare policies in the world. So under this scheme, the children will be given a certain amount at birth and they will take control of this amount of money when they turn 18. So there were multiple objectives designed for this policy. And the aim is not just to provide the kicker funds for education. Okay, Simon Kursos here, no? Oh, here. Okay, so the aim is not just to provide the kicker funds for education, housing or starting businesses, but probably more important is to help people understand the benefits of saving to develop saving habit and financial literacy so that they can make better financial decisions. So the Child Trust Fund was designed in such a way that for children who were eligible for this scheme, their parents and guardians were given a voucher of 250 pounds from the government to place with the Child Trust Fund provider, such as some banks and beauty societies. And the amount of the voucher would be doubled for some families with lower income. Apart from these contributions from the governments, the parents and wider families were also encouraged to add on to their account with a tax-free saving allowance of 1200 pounds per year. So if you take advantage of the full allowance each year that would add up to 21,600 pounds when the children turned 18, which is a considerable kicker fund for the children. And finally, the money will be available for the children at the age of 18 when they can obtain the control of the money and choose whether to withdraw it to spend it or to reinvest it. So this scheme was introduced under the new labor in 2005 and was backdated to the children born, the children who were born after September 2002. However, in 2011, the scheme was scrapped to cut the government expense in the wake of financial crisis and was later replaced by junior ISIS. And in September of 2020, the first Child Trust Funds matured and those who turned 18 then had the control over the money. So given all these benefits came with the Child Trust Fund and the government's efforts in promoting savings, the result after 18 years and not as exciting as expected. According to the data we get from the HMRC on the market value of Child Trust Fund, we find that over 86, I mean, the distribution is highly skilled to the right and about more than 86% of their account have less than 2,500 pounds saved in their account and only 0.4% of the account has like over 20,000 pounds saved in it. Therefore, we'd like to understand what has happened to the Child Trust Fund, whether or not it has made any difference on the level of savings and saving behaviors, if yes, to what extent and who benefits the most from the scheme. So there are several research questions we'd like to touch in this presentation. First, we would like to look at whether the parents are aware of the Child Trust Fund and also whether their account was opened by the guardians themselves or opened by the HMRC on default. These two questions are quite interesting because if you are not aware of the account or didn't have much motivation to open an account by yourself in the first place, it's very unlikely that you make any further contributions and you may even lose connection to their account, which means the children won't be able to collect the money when they turn 18. And then we would also like to know whether the parents made additional contributions and the amount of savings in the account and in other places for the children. And on top of this, we would also like to know if these behaviors vary across groups. So the data we used for this analysis was collected from the first six waves was an asset survey. This is a large-scale longitudinal data set and the first wave was conducted in 2006 and 2008 which covered over 30,000 households. So the last question now was asked to the adults in the household, but since our analysis focused on the dependent children aged up to 18, so in our analysis, we have to convert their original data into a child-level longitudinal data set, which means now the dependent children becomes the unit of analysis and the information on their parents and household is matched to each child. We ended up with over 60,000 observations, child-year observations for analysis, which came from over 30,000 children and 820 of them took all the six waves of surveys and another over 15,000 children were only surveyed once. So let's start from the first research question. We would like to know why the people, parents are a wire of the child-transformed account and we find that in our sample, around 15% of the eligible group seem to be under wire of their account. So this data is, well, we got this data based on the sample who was surveyed by WISE for the first time because we expected those who were surveyed for several times by WISE questionnaire will become a wire of the child-transformed account during the process. So this data is got from a sample who was surveyed for the first time. And we know that the unawareness of the child-transformed will lead to, for example, missing the benefits of the tax-free saving account as well as an issue in collecting the money when the children turned 18. Actually, the account-tracing issue are significant according to our conversation which stops from HMRC and it is estimated that more than 1.8 million child-transformed accounts are lost to their owners either because their accounts were opened by HMRC on default or that the parents move and didn't keep in touch with a child-transformed provider. And also in our sample, we find that around 70% accounts were opened by the guardians themselves and this proportion is higher among the children whose parents have graduated degree or the children who came from families in the top third of earnings. Additionally, we find that 27% of their accounts has received contributions in the previous two years by the time of the survey. And again, we find this proportion to be higher for children who has graduated parents or for children who came from families in the top third of earnings. Then we will look at the levels of total savings for children and we compare between the group of children whose parents report to hold a child-transformed which means they are aware of the existence of child-transformed account. We compare this group with the children who are too young to be eligible for child-transformed and those who are too old to be eligible for child-transformed. So it's not surprising to see that the children who are too young to be eligible have much lower savings compared to the other groups. And the reason might be that they didn't have much, they didn't have much time to accumulate the wealth. However, if you look at the graph on the right, which is the graph for those who are too old to be eligible for child-transformed, you will still find the saving level to be lower than the child-transformed group in the middle. So if we focus on the child-transformed group only in the middle, you will find the bottom quarter had done little to add to the government contribution and a slight increase from 250 pounds to about 380 pounds in the savings might mostly attribute to interest growth. This implies that they might have either lost their accounts or just were not capable to make any further contributions. And additionally, you will find a widening distribution in each of the group and the widening gap between the eligible group and the other two. However, these statistics can only take us so far because we didn't take account of the other factors that might be relevant in this context. So in the next step, we estimate a multi-level mixed-effects linear model, which allows for the fixed effects at the children's level and the random effects at a household level because we expect the children from the same family share some similarities. And we control for a set of standard control variables such as the age and gender of the children and other variables on the parents and the household as well as a year and a regional dormitory. And we summarize the key results from this model in this table. A dependent variable is a total levels of savings, which includes the savings in the Child Trust Fund account as well as the savings saved in the other places. So we find that the eligibility of the scheme is associated with 426 more in the total savings for the children. This is only a small amount if you compare it with the government's contributions. And focusing on the key control variables, we find that those whose parents were, compared to those whose parents were married, the children whose parents are cohabiting or single tended to save less for them and compared to those whose parents are graduate, those whose parents hold lower qualifications or no qualifications at all, tended to save less for them. And also compared to outright owners, those parents who are mortgagees or renters tended to save much less for them, especially the latter. We also find that children who came from families with higher wealth and income have more saved for them and those who come from families with heavy debt burden tended to have less in the savings. The results on these control variables highlights the cohorts of children that cause for more attention or support from the government. And besides these baseline results, we also did a set of heterogeneity analysis and we find that the child trust fund boosts children saving the most for children who aged between zero to four or have single parent or from families with low incomes. So if you combine these heterogeneity analysis and the baseline results, we will find that while those who come from families with lower socioeconomic status tend to save less, the child trust fund is somehow more meaningful for these children. So here comes our conclusions. So first time analysis shows that the child trust fund has increased children's savings, but in a limited way the improvement in total savings seems only a small amount which is 426 compared to the initial contribution from the government, but put that in the context that 11% of young adults have no savings at all and 28% have less than 100 savings for them. The child trust fund at least ensures a minimum level at the age of 18. And the second conclusion is that our analysis show a rather complex overall picture of child trust fund. So on the one hand, the scheme has the greatest impact on those who are more disadvantaged, but many of these families are also less likely to be aware of the account and also less capable to save. On the other hand, the rich and the better educated families are more likely to save and to save more, therefore reaping the greatest benefits from it. This is quite common for tax privilege saving policies which always skewed to those who are more capable to save. And we have to say that it's hard to find the solution to this paradox, but at least improving the awareness of the scheme in the first place and especially for the disadvantaged families would be a vital step. And finally, though the child trust fund improves children's savings, the amounts are unlikely to be life-changing for the vast majority. It's only like just over 400 pounds. Therefore, why the other objectives that we mentioned such as building up the saving habits and financial literacies, whether these objectives have been achieved becomes especially important. Just like the old saying that if you give a man a fish and you feed him for the day, if you teach a man to fish, then you feed him for a lifetime. So this calls for special attention on the delivery of complementary policies to the child trust fund scheme such as the financial education policies. And we would be happy to see more data and research on how the young adults use their accounts after they take control at the age of 18. Well, in a nutshell, the child trust fund is a great pioneer of asset-based welfare policy. It has made a difference, though maybe limited and skilled. But we still learned lessons from it for the future development of asset-based welfare policies. I guess I should stop there. Okay, thank you. Thanks, Lynn. Really interesting evaluation of that policy and lots to think about whether if it hadn't been disbanded and the promotion materials and so on had continued, then I wonder if the outcomes would have been different. But it's really good to see how that's all shaping up now and hopefully anybody thinking about policy in that area can use the results to design something that potentially works better, especially with the cost of living, meaning that people saving on behalf of their children is going to be even less like. Now, last but not least, we have B. Boileau. B is a research economist in the pensions and public finance sector at the Institute for Fiscal Studies. She's currently working on projects analysing intergenerational transfers, labour market activity among older workers and retirement savings. Feel free to take over, B. Hi, so as you said, I'll be talking about intergenerational transfers and life events today. By these, I mean gifts and loans made during the lifetime of the giver. And these are opposed to inheritances, which made it there. But there's reason to think about looking at transfers in particular. So givers obviously have more discretion over the timing at which they make these sorts of transfers, which is kind of motivation to look at them separately. And they're also treated differently in tax terms. So intervivos transfers in the UK face a more generous tax treatment than inheritances do. Again, pointing to the fact that there's a different role played by transfers. There's reasons to think that these sorts of transfers are increasingly significant. So from potential givers point of view, we see rising levels of household wealth at retirement, which might mean there's more wealth to transfers or recipients. Reforms like pension freedoms also increase the flexibility with which sums of money can be transferred at various points during retirement. From potential receivers points of view, there's slower income growth over time, meaning that there might be more of a role for transfers to play. And various life course events for potential receivers are becoming increasingly expensive over time. So for example, house prices are getting more expensive. So this might point to again, more of a role for transfers to play. In this project in particular, we're seeking to answer three key questions. Firstly, we're looking at whether these kinds of transfers are growing over time in a way that might reflect their increasing importance. We're seeking to document the characteristics of givers and receivers of these sorts of transfers. And then we're looking at the potential drivers of these transfers. So in particular, we're looking at the life events for givers and receivers that these transfers might be associated with. We've been using the wealth and assets survey to investigate these questions. We've been using seven waves of it from 2006 to 2020. And in this data set, we've got questions about the size and the use of gifts and loans separately that are received. And these are gifts and loans of more than 500 pounds. And we also have more detail on givers from around seven. So this, we've got questions about giving. We've got questions about the relationship between givers and receivers. And then we've got more self-reported detail on the use of transfers, which I'll go on to talk about in more detail later in this presentation. There's been some literature in the UK context on how these transfers look. So this has largely been documenting kind of the size and prevalence of transfers and on the characteristics of receivers. And there's also been some international domination of what tends to drive these transfers, which looks at life events for receivers in general and finds that they tend to be important in checking transfers, things like marriage, divorce and childbirth. So we can look at how many people report receiving these types of transfers. And this does seem to be increasing over time in ways that might reflect the kind of increased significance that I just talked about. So you can see from this graph that about 6% of adults in the UK report having received at least one of these sorts of transfers over the last two years. But from 2017 onwards, this seems to jump upwards. And as you can see, it jumps to about 8% of adults in 2020. This is largely driven by an increase in the proportion of people who report having received the gift, which is the green line on the graph, whereas the proportion of adults who report having received the loan offers around 2% and is relatively constant over 10. So this, as I've said, might point to an increased significance. We can also look at how the value is changing among receivers to see if, as well as the prevalence apparently going up, the value seems to be also increasing. And it does seem as if the value of large transfers seems to be increasing over time in a way that might also reflect more of a role for these transfers to play in the UK. So the median, as you can see, is relatively constant with about 50% of people receiving under 2,000 pounds from 2010 onwards. But the mean seems to be rising somewhat, although this rise isn't statistically significant, it's slightly fluctuating around. You can also see that the mean is higher, not only than the median, but then the 75th centile, which really points to how skewed this distribution is and to how much large gifts are very large. We can look at who tends to receive these transfers as well, and I'm going to show you the average age of people who tend to be receiving. So this shows you the proportion of people who report having received one of these transfers over the age distribution. And as you can see, transfer peaks at younger ages with about 13% of people in their late 20s and early 30s reporting having received one of these transfers. And the proportion of people reporting reception falling after that with about 3% of people in their early 60s reporting having received one of these transfers. And this kind of points us towards what types of life events for receivers we're going to investigate. So things that like moving into home ownership, getting married, having a child that might happen at the kinds of ages where transfers are most common. And that also begins to point towards the way in which these transfers tend to be from older to younger people. And so kind of transfers in an intergenerational way. And this is again, back to how we look at who tends to be giving transfers. So here you can see the opposite pattern holding where people in their late 20s, early 30s, about 3% of people in these age categories report having given one transfer above 500 pounds in the last two years. And this rises to about 16% of people in their 60s. So you can really see that these transfers are kind of, they're taking place from older to younger people. And this is back to how we look at some of the self-reported evidence on who transfers are received from that we get in round seven. So I'm going to show you here. And I'm showing you this for gifts only, but there's a very similar pattern that we see for loans when we look at them separately. So you can see that almost 70% of gifts are received from parents with a further almost 10% received from grandparents or great grandparents. And this shows us that not only are these transfers are kind of intergenerational phenomenon, but they're happening within families. And so potentially they're a way in which economic advantages transmitted through generations. And this pattern is kind of even further confirmed when we look at the proportion of gifts value represented by different sources, which I show you here. And you can see that almost 90% of gift value is received from parents, really pointing towards the importance of parent-child's connections in determining the nature of these transfers. So one further feature of transfers I'm going to present before I go on to talk about our primary analysis of life events, which is the fact that these transfers tend to be received infrequently. So this shows you among people who report having received at least one transfer and to a presence in all six of the waves that we look at, how many waves did people report receiving a transfer in? And you can see that most people report only receiving a transfer in one wave, with a further about quarter of the people reporting having received the transfer in two waves, but almost no one receiving them in more than that. And so rather than being a kind of continuous flow from parents to children, these transfers seem to be more of an infrequent or one-off phenomenon, which further stop our decision to look at life events as something that might prompt these large sums of money to be transferred. And so I'll go into some of the life events that we look at in particular as I go on. So I just want to talk a bit about what in particular we're doing. So because the World Nasset Survey is a panel dataset, which follows the same people over time, we're able to construct measures of life events, measures, so people's status changing between waves. And we test whether those status changes are associated with your probability of receiving or giving a transfer. And for example, we look at getting marries between waves for receivers or becoming weathered between waves for givers, among other things. We run various OLS regression specifications, which test the association between whether you receive a transfer, a vector of kind of demographic characteristics, because you might think that some people are both more likely to receive transfers and more likely to experience life events. We're controlling for that. And then we control for various life events that I've just shown you in a staple format. We also examine the amount received as an outcome, and we run an equivalent regression with giving as an outcome to test whether various life events seem to be significantly associated with a probability of giving a transfer. So here I'm going to present some of these results. So you can see that having moved into home ownership between waves, compared to someone similar who hasn't moved into home ownership between waves, seems to be associated with a 10 percentage point increase in your probability of receiving a transfer, pointing towards the importance of home ownership as a driver for receiving transfers. Marriage also seems to be another important driver, increasing the probability of receiving a transfer by nine percentage points. Moving region between waves and moving into self-employment from employment between waves also seems to be significantly associated with the higher probability of receiving at least one transfer over the last two years. And separation also seems to be significantly associated with the higher probability of receiving a transfer, although this is only just and there's a smaller effect than the others. For these next three events, the connection is not statistically significant. And this is potentially interesting because the events most connected with the higher probability of receiving a transfer are potentially more possible among other things. Whereas stuff that's less predictable like experiencing a decline in employment earnings or experiencing unemployment don't seem to be associated with transfers. And this kind of helps us to understand the nature of these transfers. So maybe they're not functioning so much as family insurance, for example, maybe they're a way of transmitting economic advantage for generations. And we can also look at the size of transfers. So this is among all adults. We can look at whether experiencing one of these events between waves seems to be associated with receiving more and transfer value than someone similar who didn't experience one of these events. Moving into home ownership seems to be associated with receiving about 2,500 times more in transfer value compared to someone similar who didn't. And this could be acting through two channels. So one which I just talked about is the fact that you're more likely to receive a transfer if you've also moved into home ownership at the same time. And another one which I'm going to talk about in the next slide is that among receivers, you could be receiving more if you've moved into home ownership compared to someone similar who also received transfer but didn't move into home ownership. Marriage also seems to be associated with receiving slightly more in transfer value compared to someone who didn't get married. But then none of the rest of these events seem to be significantly and positively associated with receiving more in transfer value. Now, when we look only at those who've received a transfer, you can see that home ownership, people who've moved into home ownership between wave compared to people who've received a transfer but not moved into home ownership between wave seems to be receiving around 10,000 pounds more in transfer value. So as the other two sides have shown, this really backs up the importance of home ownership and driving really large transfers, not only transfers at all but really large amounts of money. None of the rest of these events are significantly associated with receiving much more in transfer value than other way similar people who haven't experienced one of these events. So now I'm going to show you some self-reported evidence from round seven which kind of backs up some of the stuff that we've been looking up at so far. It's useful compliment because it allows us to look at life events that are harder to get that in the way that we've been doing it so far. So for example, whether someone reports using a gift for living expenses or whether they report using it for holiday, which we don't really have the variables to get that in was. And so this shows you the proportion of gifts used for various different self-reported reasons in round seven. And as you can see, property purchase seems important with 23% of gifts received used for property purchase or improvements. But there are other important uses that come through her too. There are many gifts used for living expenses, many are saved or invested. The others are used for holidays, for vehicle expenses or for other reasons. But when we look at the proportion of gift value represented by various different uses, property really stands out as being very important. So here I show you a pie chart that represents the proportion of gift value used for various different reasons. And as you can see in more than half of gift value was put towards property purchase or improvements. And this is another way of kind of backing up what we've seen so far, which is that property purchase is really important in prompting large gifts in particular. And we see a really similar pattern when we look at loans, which I'll show you now. So again, when we look at the proportion of loans used for different reasons, property is important, but as our various different reasons, like debt repayments, again, living expenses and vehicle payments, among other things. But when we look at the proportion of loan value used for different reasons, property again represents more than half of the total value used. Finally, I'm going to talk about some life events for givers that might be associated with a higher probability of making a transfer. And this is to try and kind of get out what's driving some of these transfers from the giver side of things, whether these gifts are solely responsive to events like a harmonious ship for receivers, whether there might also be some events in givers' lives that front transfers. Having been withered between waves compared to someone otherwise similar who has not been withered between waves is associated with an almost 15 percentage point higher likelihood of making a transfer over the last two years. And having inherited also has a statistically significant association between with making a transfer of over 500 pounds, increasing the probability by six percentage points. These next three events, though, don't have a statistically significant association between experiencing those and making a transfer. And this is quite interesting because unlike what we saw for receivers, being withered and inheriting seemed relatively different to foresee, especially when compared to the other three events that we look at, which don't seem to be associated with a higher probability of making a transfer. So whereas foreseeable events seem to be driving transfers for receivers, this doesn't seem to be so much the case for givers. So I've shown you that transfer value and transfer prevalence seem to be slightly increasing in recent years in particular, which might reflect some of the reasons to think that these transfers are playing an increasingly important role in people's lives. Talked about the kind of relationship between givers and receivers, which tends to be parent, child and nature, and the fact that these transfers are working on an intergenerational within family basis. Talked about the events that seem to be associated with transfer behavior, so becoming a home ownership, marrying and becoming self-employed in particular. We've seen that home ownership is particularly important in prompting very large gifts. And then for givers, it seems that events that might prompt transfers are less predictable. So things like inheriting and being widowed, whereas we don't see a strong association with other events that might change how people are holding their wealth, like drawing a pension or paying off a mortgage. And that's it for me. That's many questions.