 to our next session. So we're now going to hand over to the DWP to hear their latest updates. So this includes the Family Resources Survey, Households Below Average Income, Income Dynamics and the Pensioners Income Series Statistics. And the presenters are Mark Vaughan, Natalie Lloyd, Lorraine Pearson, Elias Imgy and John Bilestone and Helen Smith. And they're all from the Surveys branch of the Department for Work and Pensions. Okay, I'll come on camera. Thanks so much, John, for introducing us. Yep, good afternoon, everybody. Really wonderful to see so many people attending the annual conference on these servers yet again, this year. Really good attendance in recent years. And I hope you're taken away from today, just as much as I am. I think that's some fascinating talks already. Today, the material that I have for you will have links to the kind of concepts that Peter Matthews was just discussing. And also reaching back to this morning a couple of links as well to the really good talk on admin-based statistics on income and census that we had from ONS. So I did enjoy that. If I can have the next slide please. So yep, good afternoon once again. My name is Mark Vaughan. I am the Project Lead on the Family Resources Survey based at DWP. We have a couple of links on the title slide for you. The first of these is our homepage at gov.uk. The FRS is a published national statistic. You can look on gov.uk for the headline stats. I'm quite happy. So why not do so? And secondly, we do have a team inbox as well for inquiries. This is attended. It is something that if you send an inquiry in, we will do our absolute level best to get back to you in a reasonable period of time. So we've got that on the slide just there. I'm going to talk about a few things today. As we usually do, we're going to say just a few brief words about what the FRS is. Thank you. A few brief words about what the FRS is. Give an update this year. Highlighting a few developments and also wouldn't be a statistics presentation without some statistics. So we have a slide on that as well. I'm also going to go on this afternoon to talk you through why the FRS is heading over the next few years and a few points there around that happening date or agenda, as I say we've heard about that already today. So a couple of points just to get across in terms of where the FRS is now, what is it and where can you access it. So I think the key things to get across would be the latest FRS for the 2021-22. The year was published in March. We published in March this year. You can access the FRS dates, it's yourselves. They are available by UK debt service. A massive thank you to colleagues at UKDS as well. This year we had the dates that's released. I think I'm right in saying within about five weeks or so. Publication day, which is a latter-day record short amount of time. It's really, really good to see that going out. So it's accessible just there. I should also advertise that it's accessible by the ONS secure research service. Now this is a brand new thing for this year. So we do absolutely welcome people's proposals for research by the SRS at ONS. The other thing that is brand new that we really need to highlight is static explore. If you've never come across to explore before now, this is the department's online tabulation tool. So it does enable people to create their own statistics, create your own tables in an Excel style format and your own breakdowns of data. Static explore is not new in itself. The department has had static explore now for a long, long time, so well over a decade of operation. But historically it was always facing the department's admin data systems by which I mean, how many people in total do you have receiving state pension? That's the kind of question that you'd answer. And there you'd be looking at the population as read-in from administrative systems. Developments in recent years, we've had HBI and pensions incomes for just a little while now. I'm really pleased to say that FRS has joined that group. You can now do your own tables of results from the FRS data sets using static explore. And we are very conscious that for people who perhaps don't want to launch into the micro data but do have a need to get some headline statistics or key breakdowns, this can be a very valuable tool for them in allowing them to do that themselves. So that's a major development brand new this year and a massive thanks while I'm here to both Justina Owen and also Anna Britton in our team who pulled that together for us. Moving on, the FRS, if you want to wear this is one of the country's largest household surveys, focuses on incomes at the individual and family and household models. The FRS is a very, very detailed study. It looks at income from a very wide range of sources, everything around working earnings, but also pensions and payments for those who are retired. Obviously the world of state benefits, we are deeply pleased that there's a massive focus on that too. And it does intend that there are other fringe incomes that the survey reports as well such as investments and annals rental profits. Survey also has a very wide range of characteristics. We have a lot of the demographics that you might expect, but we also run through the world of housing and tenure. We do quite a lot in the way of disability health and care. And it doesn't stop there. We do a lot of other things around education, people's education levels, and also child maintenance as well, which again is a natural focus for the Italy P as a department. There are other facets that we look at. I would probably highlight material deprivation. We're also branching into food security and food related deprivation. More on that in just a couple of slides. Time. In terms of the content of the interview, we have been at an hour long interview on average for quite some time now. The questionnaire content is essentially unchanged from one year to the next, but we do make some changes as the years go by. So that's just a bit of backdrop and a couple of the key points. If you take nothing else away from today, those will be the key points. Next slide please. Thank you. Developments this year, as last year's conference, we've divided these into the red group and blue groups. The red group are the things that we have under the hood, things that we can see in the blue group, effectively the things that all of you can see. We do highlight the red group. A lot of work goes in behind the scenes to make the FRS a reality. Work on any given FRS release is normally beginning in the January, late January or February, the year before publications. So a lot goes into it. Under the hood this year, I would put in place a brand new data management solution. This supersedes a solution that's been with the department for towards 20 years. It brings us up to modern HTML-led standards. It has a whole bunch of other things besides in the way of improvements. Again, recognizing that there will be lots of people on call who won't be too concerned about that, but it's a massive thing for us and really cements the future of the FRS in data processing terms for many, many years to come. That's all good on that. Again, under the hood we've got several and call it checks this year. We are always on the lookout for ways that we can improve the quality of the FRS, albeit it is a national statistic. We are always looking across to the code practice and the obligations there where it comes to continuous improvement. So quite a few things that we have done. I would single out probably state benefit changes there in Scotland as something which we do have to look at quite closely just there. That will become sure an increasing thing for the survey in future years. Thirdly, in that group cost of living, this is something that we are having to consider quite carefully in terms of how we class it as income and what do we do with it, the kind of thing that's something Peter during his time with the department would have given very great thought to, I'm sure. A lot of thought going into that. In terms of other changes though, which I'm sure you're more interested in, things that you can see. There's a few to call out here this year. So we've got things like within the Excel tables that we publish. We've got things like the inclusion of a new variable, a new category for sandwich carers, those who are caring for both kids of their own and also somebody else and also parents. We've got a brand new variable for those on zero hours contracts. That was in response to user feedback. Third point is a little bit of a techie one, but it's around the way that we classify those working in the public sector. And the driver for that really was to bring the FRS fully linked up and locked in to the world of public service pension schemes. A couple of new variables alongside that, one called Doc to one called Dentist and a few others besides about Docs and Dentist. It was put to me that the FRS was simply being curious about Doctor and Dentist earnings. I can assure you that it's not the case. The reason we're doing this is because of the sheer difference both in scale and often in nature of the way that doctors and dentists earn money and the way that the FRS handles those individuals. It's a very, very key change there for us looking at doctors, BU working in the NHS, BU, a consultant to register in a hospital or BU a GP. We have a marker for that in the survey now and I'm really pleased to see that as an improvement in that quality. I think somebody this morning on the the fifth bullet mentions looking at using the internet at home for online services. This isn't a binary question yet to know that it is nuanced in terms of how much support people think they might need to use internet services in the home and against really interesting findings coming out of that. Second to last is food bank usage. I'll come back to that in a second, but finally I really do have to highlight that if you are interested in what will be happening on the FRS going forward, you can go to a homepage at gov.uk, you can look at what we are calling the release strategy for the FRS and that will give you some good information on the developments that we intend with the survey this year, this publication year and indeed next year as well. So it does get on to things like changes to questions that are going into the fields and those sorts of things. Next slide please. Okay so at the top it wouldn't be an FRS presentation without some statistics. Here they are. Possibly the most high profile changes this year that we've just published was around food banks, so we did have a question in the fields. Have you used a food bank? When we put that into the field, the accent there was have you used a food bank effectively for emergency purposes? Had to get a hold of foods on an emergency basis. As the results came back, why were we doing that? We talked about a household food security of course. We put that into the fields just a little while ago and really this is all building our evidence on deprivation in the food sense of the word. So we're talking about food security which covers concepts such as affordability, worries and concerns and our people skipping meals. We're talking about food banks which is obviously very direct form supports. The direction of travel here is that we will look at is anybody else feeding you as well. In other words, we don't mean people bringing around small volumes of food as present. We mean relatives perhaps bringing material food supplies around to you to support you. So we will be taking the survey in that direction as well. When it comes to food bank usage, we asked have you used a food bank in the last 30 days? We also asked whether you'd used one in the 12 months running up to interview and the stats are on the slide right there. So the finding was that we had one cent used in the last 30 days and three percent of the cost last year. As you might imagine, it's possible to break into those figures and look at things like regional variations. We've got a map on the left hand side which does indeed show differences by geography. This is by region. If you were to look at the FRS publication on gov.uk, I think have around 10 Excel tables which give a variety of other breakdowns and cover things like the extents which are lower incomes associated with high food bank use and a range of other factors besides as well. I'd probably better move on to the next section so I can have the next slide please. Okay, so I am not Don Burke. I'm presenting slides here today that Don has put together and this is really trying to get across where the survey is going in relation to greater use of admin data. So if you take the next slide please, thank you. The first slide we've got is talking about the rationale. A couple of speakers have covered this kind of thing already but it really does fit in very well with that theme of making much greater use of admin data and leveraging the value from those sources. In terms of why we're doing this sort of the rationale, there are a number of drivers, not least improving data quality. State benefits do vary in the extent to which the FRS has an undercount. State pension are quite close for other benefits. We do have material undercounts but one of the things we're trying to get at in including admin data is closing that gap if you like of this is what the survey says, overhang on, that's different to what the admin source says. So there is a point there around consistency, coverage, accuracy, call it what you will. There's certainly a point of timeliness. Again, I'm thinking about what Peter was talking about just there. The long-term game plan here is that if you can include more and more administrative data on things like state benefits and other sources as well, you will have to do less and less in the way of the current editing and checking that we do of FRS responses and that should other things being equal make the publication, the release of the survey data set much more timely, much closer to the point of interview. And third point obviously is that if you reduce time you can also reduce cost. The ultimate game plan on this is that you will be able to reduce the length of the FRS questionnaire from that one hour mark that it currently sits at. So three big points there for the future of the survey. Can we do this? Yes, we certainly think we can. We have and have had the ability to link respondents to their administrative records several years now, thanks to a change in the legal basis for consent. The latest that we have on this is that our match rates, the number of people who can successfully look up to think that is up about 95% respondents. That's pretty good. That's pretty good. So is there a good reason to do it? Yes. Can you do it? Yes. If you are going to do it, what are the main research strands? I think there's a few here on the page where we're talking about let's look at the sources. Which sources can we bring in? So things like state benefits, obviously we as DWP sit on top of the mountain of admin sources. Things like real-time information, things like self-assessment data, both of which would be HMRC concepts. Definitely giving a lot of thought to those as well. It doesn't end there. There are a couple of other things. We need to look very closely at what's happening on grossing weights when you bring admin data in. We also need to look quite closely on what it means for non-response bias. Really, that's how different are the people who have taken part in the survey when compared to those who did not. Next slide please. Thank you. I'm not going to go through this one word by word, but if we're talking about the previous slide with the rationale, this one really is what do we actually intend to deliver from this strand? And we've got a couple of big, big, big points here. So is this going to be some kind of clean cutover, day-night cutover, where we suddenly transit to a new basis and that's that? Well, now that's not the idea. Projects are very clear that what we want to do is actually deliver some degree of back series of FRS data sets which do have this admin data in and therefore give people a chance to compare the difference perhaps that having the admin data has made to the results. So that's one big thing in terms of deliverable. The second thing very naturally flowing from that is if you do have admin data, what kind of difference does that make to the statistics that you're putting out? And that's huge for us. We will be looking for a really clear demonstration of the difference that admin has made to our statistics. And the third thing, which is possibly a little bit more internally facing but none was quite important, is to consider what this means in system terms. So if you think of the FRSC production line, what changes do you need to make? If you are reaching out to admin data and becoming reliant on that, can you still be reliant on that in two, three, four, five years time or will the admin data have evolved to a different place? It's absolutely vital we have that understanding as well. I think that's nearly it, but I was just going to close with the last slide of that, because I'm sure the question you'd all like to ask is, you know, that's what we're all about, but why is this happening? The slide that Don has put together looks like this. What is this showing you? This is showing you the next five publication years. As we said earlier, the FRSC normally published in the month of March. What are we talking about here? We're talking about a journey from left to right as we progress towards 2027. And just with that little change in 2027, you may notice that instead of March, we're proposing potentially autumn. Could be the month, could be the time of year in which we publish, reflecting the points made earlier around timeliness. Several points on the page here and I'm not going to step through all of them, but what I think is safe to say that looking at March 24 on that left hand side, what we're not proposing is that somehow you just drop the existing FRSC and it's related to publications. I think the idea is very much that we still do that, but that alongside that you're talking about an experimental rival set of statistics, which give an idea, as I said on the previous slide, of the kind of difference that having an admin led FRSC and an admin led set of HBI statistics could make. That March 2024 dates the key one that's in all of our minds at DWP right now. And we're still considering that very closely. We're also considering alongside that what does it mean for the longer term and the lower sections of this slide are talking about what does it mean for information collection? When will you see a change to the FRSC questionnaire? I think that's probably about my time. So I'm going to draw to a close just there. If you have any questions, do pop in the chat to me themself that team will try to pick them up. But that aside, thank you very much for listening. I'll come straight up to Natalie Lloyd who'll take you through HBI. Thank you. Thanks Mark. Hi, for those who don't know me, I'm Natalie Lloyd and I'm the lead statistician on the HBI publication. I'm sure I'm assuming that most of you on the call are familiar with HBI, but you certainly will be having seen Peter's presentation. So I want to send some thanks to him actually for covering some of the key trends in the statistics over the last few years. But in a nutshell, the HBI publication is the official source of UK poverty statistics. So we publish estimates of the number and proportion of individuals, pensioners, children and working-age adults who live below the poverty line or they are in low-income households. I don't have a great deal of time here today. I've only got a sort of a 10-minute slot. So I'm going to use the session just to really talk through the last publication and the changes that were made there. And then also a bit of a forward look for the work that we're going to be doing over the next few months and beyond just to bring you up to date with everything happening in the HBI world. On the last slide I did leave an email address, happy for anybody to contact me with any questions after this meeting and I will try and answer anything you put in the chat too. So thanks for that. Yes, you can move on now, thanks. So those of you who may have attended last year, if you can cash your mind back to that or if you're familiar with the publication, you may be aware that when we published the 21-22 statistics back in March 2022, we had kind of, we spent a lot of time, invested a lot of time in understanding the impact of COVID-19 on the results that we were seeing in our estimates. So due to the move from face-to-face interviewing to telephone interviewing and the reduction in the sample size during the 2021 survey year, that presented kind of a lot of problems with the underlying data. We had to do a lot more validation of the data than we ordinarily would to make sure that the estimates we produced were as robust as possible. What this meant was that we didn't publish our full suite of statistics because at a kind of a lower level we couldn't validate all of the trends that we were seeing and we then accompanied our publication by a technical note for users to offer full transparency over why we took the decision not to publish all the statistics and the kind of areas where we would suggest kind of key caveats should be applied. I'm pleased to say that in the 2021-22 publication that we published in March we did return to publishing our full range of statistics including a full update of stat explore. We did produce a technical report once again although it was far less bulky than the year before. We made the assessment that although there was still an effect of the change in mode it was affecting the estimates to a much less of an extent than in the previous year. So just as an example we were seeing previously that following the move to the telephone interviewing we were getting kind of more older participants responding to the survey tended to be more affluent, more educated and we saw all that kind of coming through the results. Although we did see shift in the sample representativeness continue into 2021-22 it was much less marked so we were much more confident in kind of issuing our full release of statistics. A couple of things to note though what we did do was we didn't retrospectively publish the 2021 data so that's still not publicly available although it can be obtained by accessing the data sets in the UK data archive. In terms of our published estimates for all those estimates where we would use a three-year rolling average so that tends to be around ethnicity and regional estimates we had to use the two data points we had available so we excluded the 2021 figures from the estimates. There were a couple of additions to our publication this year so we had previously published back in March 2022 some experimental statistics around combined low income and material deprivation rates for working age adults. We brought that into HBAI this year and as Mark covered in his section there were some new statistics on food bank usage within the FRS and so we built on these by looking explicitly at food bank usage among individuals who were living in low-income households so that's all in there. Move on please, thanks. So in terms of our future plans and where we are now as we speak actually we are starting to look at the 2022-23 FRS data so we've taken delivery of the six-month dataset covering April to September of last year and we're making a first assessment of that data looking at the impact of the cost of living pressures on household incomes and getting first glimpse of the low income rates and impacts across the board. One thing we're really going to have to focus on this year though is making an assessment of the effect of having a mixed mode during the 2022-23 survey year and what impact that has had on both the FRS sample but crucially what effect that's had on the HBAI estimates. You know in the past we've hypothesised that there will be some effect of having a different mode and this year we can quantify it and measure it and we can look at that split by different subgroups for example. So we're very much planning to look at that in detail and you'll consider what adjustments if any we would need to make to our data and our estimates to reflect the mixed mode so that that's ongoing. Again following on from Mark's presentation we will be investing a lot of time into quality assuring the integrated FRS data so that's the FRS linked with the DWP benefits data which we announced back in March that we would publish something showing the kind of illustrative value of that next March. So we've taken delivery again of some of the back series of that data that linked data and we're looking at that in detail and we are having kind of discussions internally around what would that look like in terms of a publication. You know will it be as extensive of what we currently produce for HBAI or will it be much more limited and what precisely will the publication look like. You know how we've kind of present our findings and trends etc. So yes in the short term there will be two publications running in parallel for several years but we still got to work through a number of making a decision internally about what that actually means and I saw that in the chat although I haven't had a chance to answer the question yet somebody did ask about the new material deprivation measure that we have been working on so you may be aware that we've been working in partnership with London School of Economics on refreshing on material deprivation measure including refreshing the questions that we ask. So the LSE in the kind of stages finalising their recommendations there and we will have a little bit more work to do on the methodology but the the new questions have been introduced in the field this year so they've been kind of you know being part of the survey since April so it'll be the March 2025 publication where we will be moving to using a new material deprivation measure and I'll try to answer any further questions on that in the chat. Just briefly then you will have seen probably when we published the data in March we accompanied the release with a statistical note which confirmed that as a department we were going to resume work on developing a new measure of poverty using the Social Metrics Commission recommendations and methodology that was kind of made I think several years ago. So I'm now going to hand over to Lorraine Pearson who's leading on this work to tell you a bit more about that. Thanks very much. Thanks Natalie. Hi everyone so I'm in a slightly different team in DWP I'm in the income analysis team but work very closely with other DWP colleagues on the call. Our team inbox is shown on this slide and my email address so if anyone has any questions or wants to follow up afterwards then please do get in touch but as Natalie said today I'm just providing a very quick update on work that we're doing to take forward developing new experimental statistics on poverty measurements so if we can move slide please. Yeah so back in March we announced our plans to resume this work which will be to develop an experimental measure of poverty based on the innovative work of the Social Metrics Commission so as I'm sure many on the call are already familiar the SMC are an independent commission and they were formed with the aim of developing a new approach to poverty measurement to better reflect both the nature and experiences of poverty for different families in the UK and they had a key aim of building consensus around poverty measurement in the UK as well so they published an initial report in 2018 setting out their methodological approach and they've subsequently published updates to that with updated analysis and some methodological refinements as well so if you're not already familiar you can find all of that at the Social Metrics Commission website which is the link is on the slide here if we can move slide please so just a very high level to set out what the poverty measure is so the SMC measure is consistent with HBAI in terms of its measurement of net income but then it adds additional components onto that so it seeks to take into account the availability of liquid assets and to also account for what they term inescapable costs so taking into account debt that people face well families face child care costs and the costs of disability as well as some recurring housing costs not currently accounted for and they also seek to include groups that have previously been omitted or underrepresented in poverty statistics that rely on household surveys so there's some discussion around the rough sleeping population and around housing adequacy particularly consideration of overcrowded housing and then some of this is familiar from other presentations today but they set out sort of different dimensions to the poverty measure to include looking at the depth of poverty so the distance from the poverty line according to this measure poverty persistence so how long people are spending in poverty and moving in and out of poverty and then they also look at what they call a lived experience indicators so basically what characteristics are associated with the experience of poverty using this measure next slide please so in terms of the next steps for DWP we accept that this approach adds new insight to our understanding of poverty and we'll be taking this measure as a starting point to assess how it can be developed to increase the value to the public of these statistics we'll be looking at how we can realise some of the concepts in practice not all of the data is necessarily available and there's kind of methodological decisions to make so we'll be looking at all of that we're currently at a very early stage we're at our initial planning stages we are planning to publish a series of experimental statistic publications and we're establishing stakeholder groups at the moment to see this work so you'll be hearing more in due course we'll be updating via the DWP statistical work program and sort of wider stakeholder engagement and public consultation as well so it's just a flag at the moment that this work is starting definitely happy to take any questions and to follow up with anyone who would like to know more but we will be putting more information out on this soon as well but that's it for now I'll hand over to our next speaker hi everyone I hope you can hear me so I'm Alia I'm the pensioner's income publication lead for the 2022 publication published in March this year I'll quickly just talk about who we are what we are and some key results from our recent publication and then I'll pass over to John who will talk about the income related benefits estimates of take-up publication next slide please Kate pensioners income so who are we we're an annual national statistics we report on public sorry we report on pensioners incomes from the family resources survey and our statistics examine how much income pensions get each week and where they get that income from we look at how incomes have changed over time and variations in income between different types of pensioners our estimates are normally based on a sample of around 7,000 pensioners in private households in the UK the population age distribution has changed a lot since the start of the series and pensioners now make up a larger proportion of the overall population changes in the economy and to the benefit system mean that the amounts and components of pensioners average weekly incomes have changed over time and our statistics look at these changes next slide please Kate next slide oh yeah perfect thank you so some key results from our most recent publication so in financial year ending 2022 pensioners had an average income of 349 pounds per week which was a statistically significant increase from 1995 when it was 176 pounds the decrease from 2021 when it was 376 pounds was also statistically significant but this result does need to be seen in the context of the factors affecting the 2021 data set relating to the coronavirus pandemic so when preparing the 2021 estimates some estimates some evidence was found that the pensioners sampled were more affluent compared to previous time periods so this just means they were more likely to be owner occupiers receive an occupational or personal pension have income from investments and on average had access to higher levels of savings but the 2022 population was more closely aligned with the pre-pandemic period on those metrics so the financial year ending 2022 pensioner median incomes have shown larger decreases across most measures compared to financial year ending 2021 and so while some of this change is real reflecting the context of below inflation operating of the state pension in 2022 the degree of change is also likely to have been somewhat affected by changes in pensioner composition between the two server years so this broader context should be borne in mind when interpreting observed changes in pensioner low income rates thank you very whistle stop tour of pensioners incomes just pass to John now thank you thank you hello I'm John Bloverstone and I work as part of the team which produces the income related benefits estimates of take-up publication just let you know that this is an annual statistical publication which reports on take-up of benefits from the FRS and take-up refers to the receipt of benefits that somebody's entitled to by caseload which is obviously percentage of eligible people and expenditure which is the percentage of all money claimed and to produce this publication FRS data is matched to administrative data to produce estimates for the main income related benefits which currently is just pension age only which applies to pension credit and housing benefit and our latest publication as you can see in the slide there is that seven out of ten people entitled to pension credit actually claimed the benefit and 77 percent of the total amount of pension credit that could have been claimed was actually claimed and for housing benefit for pensioners eight out of ten of those people entitled to claim housing benefit for pensioners claimed the benefit and that 88 percent of the total amount of housing benefit pensioners that could have been claimed was claimed and just some other facts that we haven't gotten the slide there but that up to 850,000 families who were entitled to receive pension credit didn't actually claim the benefit and an expenditure of up to 1.7 billion of available pension credit went unclaimed so on average that amounted to around 1900 pounds per year for each family entitled to pension credit did not actually claim the benefit and similarly for housing benefit there was up to 260,000 pensioners who were entitled to receive housing benefit for pensioners but didn't claim the benefit and that amounted to 1.1 billion of available expenditure that went unclaimed and on average this amounted to around 4000 pounds per year for each family then were entitled to receive housing benefit for pensioners but didn't actually claim the benefit our stats obviously quite important and some main uses of those stats without the first pension credit day of action was on the 15th of June last year and that was obviously an initiative to try and increase the take up of pension credit and encourage people to claim pension credit if they were entitled and our stats were heavily used in that campaign again this year there's been another pension credit week of action which was between the 12th and 16th of June and again that was all about raising the awareness of pension credit and tackling some of the myths that may stop people applying or encouraging people to check their eligibility and make a claim and it even made the news a television program on the 13th of June where Martin Lewis did a pension special program and he encouraged viewers to check if they were actually entitled to pension credit and obviously claim it as soon as they could and now again our stats were used for that and next slide please and those last stats that have just been shown on that slide would just to say that we weren't able to publish our stats for the financial year 20 to 21 obviously this was due to ish date issues following the coronavirus pandemic we're hoping to actually publish statistics for the financial year 21 to 22 but we're currently in the process of just checking to assess that if it will be possible to publish these stats in line with the UK stats authority code of practice and obviously if it's found that we can publish those statistics then obviously we'll announce that via the stats work program put everything on the internet announcing that as soon as that's possible and currently we're unable to estimate UC take-up rates at present but we're monitoring the situation and obviously we'll make changes in the future as and when that's needed because from December 2018 there can be no new claims for any of the working age legacy benefits therefore the most recent publication for 1920 focused on take-up for pensioners only so that's a quick whistle-top summary there if anybody would like any further information then just drop us a message in the chat and we'll get back to you straight away. Hi there um I seem to have lost access to my camera so apologies for that I had it this morning but um it's no longer there so um anyway uh yeah so I'm Helen Smith and I work on income dynamics and today I'm just going to talk briefly about our recent development work and some of the headline findings that we published in March this year next slide please um thank you so um for those of you who are not aware of income dynamics it's annual statistics based on the longitudinal panel survey understanding society we produce three main areas of analysis um persistent low income rates which are measured over a four survey wave period so each understanding society wave covers two years and so it's four of those uh we do analysis on movements into and out of low income over a two wave period and we also do some longer term income mobility analysis across the whole income distribution um and over the past year we have been focusing on bringing in USOX immigrant and ethnic minority boost sample into our statistics so the um this boost the IEMB was introduced by USOC in 2014 to 15 to maintain the representative representativeness of their sample it comprised around 2,900 households from targeted immigrant and ethnic minority groups and we didn't initially include it in income dynamics because at the time of the first ID publication in 2017 there was not four waves worth of data on IEMB members to enable them to be included in the persistent low income estimates um so over the past year we've been revising all of our code to bring in the boost and test the effect on on our sample um and from the table here it just shows that um actually bringing in the sample resulted in quite a large increase in our cross sectional sample size so an additional 8,000 individuals from the IEMB itself and then um also another sort of around four and a half thousand other individuals and we understand this resulted from some correction to form a non-response and then the the effect on the longitudinal sample size is less because we require people to be present in consecutive years so in terms of our persistent low income sample the four wave sample um this increased by about 4,000 individuals in the first four years after the boost was brought in we observed only small changes really in our statistics so our single wave estimates moved slightly closer to those of HBAI um and our other statistics such as our persistent low income rates changed by very small amounts kind of in the one percentage point ballpark if at all um but nevertheless we you know we're understanding that the benefits are improved representativeness and we've also seen less suppression associated with um the smaller sample sizes and some of our breakdowns um next slide please Kate so in terms of our our headline findings um so persistent low income findings um look at who has income below 60 percent of the median for three out of four years as with the other publications produced by the branch we include before and after housing cost measures so we generally see that rates of persistent low income are stable um and apart from some small changes for for children this year um which these don't build on any longer term trend um they the rates were unchanged on the previous four wave period for for the other groups and then we also analyze rates of entry into an exit from low income across each two wave period um and we see generally typically typically sort of similar numbers of people entering to an exit from low income each wave but when we express these as rates we see that the exit rates are somewhat higher than entry rates and this is actually because the denominator for low income exits is a much smaller group because it's based on those who were in low income in the first of the two waves compared to the larger group of those who weren't in low income upon which the entry rate is based so that is reflected um in the rates here so we so it's it's not that we have that many that much more people exiting low income it's it's just to do with the denominators on which the two rates are based um entry rates are generally stable over time um exit rates uh fluctuate a little more anyway um we did see a small increase this year and that's partly explained by the addition of a new category on student intuition fee loans in the most recent USOC questionnaire wave um that we've included and that effectively caused some individuals who were in the low income group in the first of those the two wave two most recent waves to exit from it so that just helps to explain that increase there and we'll see what happens next year but it would expect it will probably stabilise a bit um and then finally um as part of our sort of longer term mobility analysis the Sankey diagrams um compare the income quintiles that people were in at two different points in time so uh for those unfamiliar with them income quintiles divide the income distribution into five equally sized groups and here we're comparing the position that individuals were in in 2015 to 16 um to where they were in 2020 to 21 um and the key message from this analysis is is is pretty much the same year on year and it is that most movement takes place towards the middle of the distribution so reflected in the sort of wider bands there um around the sort of two three and quintile two to four range and then there is less movement at the bottom um and then least movement in quintile five so at the top of the distribution um next slide please Kate got five minutes for this session yeah thank you this should only take a couple of minutes so um yeah so so finally I just wanted to highlight um a small extension to our events analysis that we brought in this year as well so our events analysis is linked to our low income uh entry and exit um analysis that I mentioned on the previous slide and it aims to dig into that a little bit more by um exploring how household level changes are associated with those uh entries and exits so in relation to low income entries um on the left hand side we have included um a change from full time to part time work in the household and in the box below that a change from having somebody in paid work in the household to having nobody in paid work in the household so each event um has three different statistics linked to it so the first statistic uh which is the four percent and the three percent on the left hand side that shows how prevalent the event was so what it's saying is that neither event was common so just four percent and three percent of those not in low income experienced this kind of change across the two-wave period um but it did increase their risk of entering low income so if you did experience this change um 21 percent of people and then 25 percent uh where there was a change to worklessness entered into low income and that compares to eight percent of all of the individuals who entered low income so we can see it does increase your risk of entering low income and then the third statistic um shows that around 10 percent um of all of those people who entered low income experienced those changes and you can experience more than one change so um that yeah that we don't see those statistics is is is adding up to 100 across across the range of other events that we produce we produce a number which are all in our publication um so when we look at low income exits on the right hand side we looked at a change from part time to full time work in the household and then we also looked at what happened if um the household experienced a change of going from a position of having nobody in work to having somebody in work across the two-wave period so again I mean neither of the events was was that common um six percent of individuals lived in a household that experienced a shift from part time to full time work and then five percent in a household that um experienced a move from being workless to having somebody in paid work but then both events were closely linked to exiting low income so um if there was a change from part time to full time work this was associated with a 69 percent chance of exiting low income and then if there was a shift from worklessness to be to having somebody in work it was associated with a 60 percent chance of exiting low income and then just as context um we know that 39 percent of all people exited low income um so just just showing then how how these actual experiences did increase the likelihood of exiting from low income compared to other people um and then yeah the they weren't experienced by you know again sort of 10 percent or fewer people who were not who exited low income experienced that change so um yeah that's um that's just a brief update on the developments we've brought in this year in income dynamics and we are in fact still working on some aspects of the IEMB integration that have been a bit less straightforward so that's our sort of current priority this year in terms of development work and that's all from me thank you