 So I'm Liz Hargreaves and Laura Kees and Fahri Brown are here with me today. We're from the FRS research team in ONS and we are responsible for delivering data to DWP on the Family Resources Survey. So today we're just going to give you an update on developments in FRS field work over the 21-22 survey year, so that was the year leading up to March just gone. So I'm going to start with a recap of field work up to April 2021 and then I'm going to give you a quick summary of operational developments. So firstly the completion of the Nocturnage rollout and then a small-scale face-to-face trial. Next I'll cover the biggest development of the year which was the launch of the FRS boost. Then I'll be handing over to Laura and Fahri who are each going to present the results from an experiment which was carried out late in 21-22 and finally I'll wrap up with a few thoughts about what we've got planned for 22-23. So in terms of what field work looked like previously, this slide here summarizes the approach before Covid which I hope most of you be familiar with. So basically we used to send a letter to respondents which included an unconditional post office voucher for £10 and the interviewer would then follow up with by visiting the address and when they made contact they would seek to interview in person in the respondents house. This slide summarizes our response to Covid which was in the year 2021 so that's the year that was just recently published by DWP. We made lots of adaptations to enable field work to continue amidst the restrictions and these included no small adjustment to telephone interviewing and various adjustments which enabled us to get contact details so tele-matching the introduction of an internet portal and the adjustment which had the biggest impact was not to nudge at the bottom of the slide. So building on the success of the not to nudge we continued the rollout into the following survey year, the 21-22 which we're talking about now. So just give you a quick reminder of what not to nudge is. I did present this last year. Basically what we did was interviewers who had been unable to make contact by the telephone were sorry for cases where contact hadn't been achieved by telephone. Interviewers were sent out to visit those addresses and if they made contact they would have a doorstep conversation they didn't enter the house and they would seek to collect a telephone number and to make an appointment to complete the interview over telephone. So as I mentioned it was launched in November 2020 and by March 2021 we had rolled that out across the whole of the GB, Great Britain. But Northern Ireland had yet to implement so they had some trial scheduled for spring 2021 but those got postponed and eventually went ahead in June 2021 so within the 21-22 survey year and as a result they were able to fully implement not to nudge from July 2021 meaning that we had full UK coverage. The impact of not to nudge as I mentioned was quite significant so in November 2020 we saw a 12 percentage point increase month on month when it was introduced in GB and in Northern Ireland we saw a similar impact with response rates increasing from 22% to 49.5%. Another operational development that I wanted to mention is a small scale trial carried out of face-to-face interviewing under post-COVID conditions. So this was undertaken by ONS between August and December 2021 and considered how face-to-face processes need to be rethought in light of COVID restrictions and changes in public attitudes. Some elements which we considered included how long interviewers should wait before entering the respondent's property, whether masks should be mandatory, whether vaccination status should be declared and how to reintroduce show cards. And as a result of the trial some measures implemented were screening questions around COVID symptoms which we interviewed or instructed to ask ahead of entering the house. Some guidance given to respondents about how to prepare the interview space for instance opening windows and minimising the number of people in the room. Interviewers were required to wear masks but it was decided that it would be optional for respondents. Similarly interviewers were instructed not to ask respondents about their vaccination status and they were advised that if the respondent asked them what there's was then they had to, it was completely up to them how they answered they could choose to reveal or not reveal. Similarly we instructed interviewers to offer breaks because obviously FRS can be quite a long interview and it was up to the respondent whether they wanted to take those breaks. Finally on show cards we took the opportunity to reintroduce show cards. So a reduced pack containing just those questions which were hardest to ask without the show cards was produced and the pack was laminated so that they could be wiped clean between use. Now I'm going to talk about the biggest development of the year which was the launch of the FRS boost. So obviously Mark already referenced the fact that we were tasked with delivering a roughly two and a half times boost. In fact it was to be precise a 2.4 times boost. So due to a couple of factors it was launched in stages. Initially funding was confirmed only for the Scottish boost so the Scottish boost launched in April 2021 and then England and Wales funding was secured later which meant that we were able to launch from October 21 in England and Wales. There were also some complications with regards to contractual limits which meant that the England Wales launch needed to be phased. So in October 21 we launched a 1.9 times boost with the full boost coming into play from April 22 so that'll be the next survey year. This slide here shows the size of the boost implemented within each of the four nations. So first you can see that Northern Ireland didn't actually have any boost as it was already sufficiently over sampled. And similarly Scotland had already got in place quite a significant over sampling. Consequently only a small additional boost was implemented there. By contrast in Wales there was no prior boost so a large boost of three times was required to support the regional analysis which was spoken about before. In England the picture is a bit more complicated as the boost varies by region with some regions receiving a similar boost to that implemented in Wales. For example, North East England and other regions receiving a lower boost which was for example, South East England and London. In terms of delivering the boost there were two main focuses for my team. First was amending the sample design and the second was ramping up the field work. So for 21-22 because by the sign of funding had been approved we'd already drawn the PSUs, the primary sampling units, sorry primary section units. The only option that we had was to select additional addresses from within the PSUs already selected so that meant that interviewers were working with very large quotas of varying sizes across the country. So for 22-23 we introduced a preferable solution which was to select additional PSUs appropriately spread out across the country in order to deliver the boost as required by region but with equal quotas of 28 addresses. Looking at the field work side, the Scottish boost was relatively minor so for us the focus on field work ramp up was that October launch. We had a big push across the summer to ensure that sufficient interviewers were trained on FRS and as a result we launched successfully in October 21 and you can see there that thanks to that boost the achieved sample in 21-22 was 63% higher than in 2021 at 16,400 households which is obviously still lower than normal but with entirely telephone interviewing was not a bad result. Now I'm going to hand over to Laura for a summary of our recent incentive experiment. Thank you Liz. So I'll start by showing you our current incentive on FRS. So we send out a post office voucher to everyone sampled on the survey and this voucher can be exchanged for £10 at any post office. So on social surveys there are two main types of incentives that can be issued to respondents. The first type is unconditional incentives and these are provided to all of the sampled addresses on the survey irrespective of whether they take part in an interview or not and then secondly you can have conditional incentives which are only issued if the address agrees to participate in the survey. So what does the literature tell us about incentives or there's quite a consistent finding in previous research that unconditional incentives are more effective in terms of boosting response than conditional incentives. However we wanted to test whether this still applies under COVID conditions because as Liz has just shown you COVID had an impact on data collection and possibly respondent behaviors as well with the other reason we wanted to explore conditional incentives is that these can have the benefit of being more cost effective. So in September and October 2020 we carried out a conditional incentive trial so this was a split sample design and we compared our standard post office voucher against a £15 high street voucher which was conditional on the respondent participating in the survey. What we found from this test is that the conditional incentive had a lower response rate than our standard post office voucher. So we decided to build on this research and carry out a further test and this time we looked at the impact of trialling a larger £30 incentive and provided it in addition to our standard voucher. This trial was run by Natsyn only during the January and February 2022 field periods. So we're going into the design a bit more. For half of the sample we offered £10 unconditional incentive which was referenced and included in the advance letter as standard and for the test group they were given the £10 unconditional incentive but they were also offered a £30 incentive on completion of the survey which was mentioned in the advance letter. So turning now to the results we compared the response rate for the control group which was post office voucher only to the response rate for the experimental group which was the post office voucher and the £30 high street voucher and what we found is that the experimental group had a response rate that was 1.37 percentage points higher than the response rate for the control group however this difference wasn't significant. So in terms of what we've decided to do with our incentives we've decided not to make any change to our current incentive strategy we're going to stick with the £10 unconditional post office voucher for now but it's possible we'll look at alternative strategies for boosting response in the future. I'll hand over to Vari now who's going to talk to you about the return to face-to-face led fieldwork. Thanks Laura so I'll start with the time scale so the intention for the rollout for the face-to-face led approach was for it to be adopted in January and the aim then was to reach 100% face-to-face led fieldwork by the start of the 2022 slash 23 survey year in April however due to a rise in cases of the Omicron variant and the associated changes in government guidance this caused a delay and meant that the launch was postponed until March and approximately half the sample were offering the new face-to-face led fieldwork approach in the first month in March with the target for reaching 100% face-to-face led now in July so I'll move on to the interview guidance now so with an evolving COVID environment changes to government guidance were implemented in our guidance of the interviewers and as Liz mentioned before the protocols adopted for in-home interviewing were developed and tested during the small scale operational trials carried out during the autumn of 2021 so in March guidance included protocol questions which recovered with the respondent on the doorstep before making a face-to-face appointment to risk assess whether the interview should take place Interviewers were also required to wear a mask in home and whereas the respondent had the choice an option to wear a mask also hand sanitizer and sanitizing wipes were available so the interviewers were able to wipe down their packs of laminated show cars between interviews and there was also other guidance as well so I'll move on to the fieldwork approach now and the instructions for interviewers so firstly we have our previous approach which is the phone-led approach and this is where telephone contact should be attempted on all of the telephone numbers available for each case and where contact is achieved by telephone the telephone interview should be offered as the default where contact isn't achieved by telephone the interviewer is to visit the address and where contact isn't achieved on the first visit they should make visits up into a total of three so that's once in the morning once in the afternoon and once in the evening and then where the doorstep contact is achieved there the interviewer should collect telephone contact details and offer a telephone interview so now I'll move on to the face-to-face-led approach that we've been trialling so all addresses are to be visited and where contact isn't achieved further visits again should be made until the total three visits have been made for those not successfully contacted after those three visits contact should be attempted via all contact telephone numbers provided to the interviewer including portal and telematch numbers and from there all respondents successfully contacted should be offered the in-home interview as default so I will move on to results now from the March 2022 trial so the March total sample was 2,542 and if I break that down into the two groups we have the phone-led cases with 55% of the sample and the face-to-face-led cases with 45% of the sample so in terms of response rate so that is complete interviews as a proportion of eligible addresses we found that the response rate was four percentage points higher within the face-to-face group and this high response may have been driven by a substantially higher contact rate in the face-to-face group but overall we found that 43.8% of interviews were achieved face-to-face within the face-to-face-led group so this may be down to respondent reluctance to allow interviewers in the house or interviewers may have lacked confidence to push for face-to-face in the post-COVID environment but we are looking to encourage for more interviews achieve face-to-face so I'll hand over to Liz now to look ahead to 22-23 Thank you very much Yeah just say I'm three minutes left That's fine this is the last slide so yeah I just wanted to give you a quick view of a few of the next steps that we have in mind so as Vorice mentioned we've begun the roll out of face-to-face so that started in ONS in March and Natsen followed suit in April ONS are aiming for 100% by this month July and Nisra will also be launching face-to-face in July so shouldn't be too long before all respondents are being offered a face-to-face interview In other areas we have seen the end of DWP telematching from this month July so that decision was taken basically because of the face-to-face roll out which decreases the need for that telematching similarly Natsen and Nisra have ceased collection of telephone numbers via a portal they've reverted to pre-COVID approaches and ONS will be considering in the months ahead whether our portal should be retained or dropped and like Vari just said our major focus now will be on how to encourage greater participation via face-to-face so we'll be exploring options to encourage respondent participation by that mode which include for instance explaining to them that the survey is optimized for face-to-face collection and encouraging them to therefore take part that way if they're comfortable and also like Vari mentioned we'll be looking at ways to build an interviewer confidence in offering face-to-face wearing mind they've not been doing it for the last couple of years many of them have joined in the last couple of years so they've never done it and it is a bit more challenging in this new world of respondent reluctance so yeah we'll be looking at ways we can encourage that so thank you for listening and welcome back I hope you had a good break we'll now move on to updates from the Office for National Statistics so we've got the household finance statistics transformation and income earnings coherence update with Carla Kidd and Ainsley Wood so Carla Kidd is the assistant deputy director for household financial statistics transformation at the ONS and prior to this she was head of wealth pensions and spending analysis at the ONS and Ainsley Wood leads on the income coherence of GSS income and earning statistics we'll then hear from Carla again with a presentation on analytical projects using the wealth and assets survey and then we'll have a presentation on the cost of living analysis with Amory De Silva so start off with Carla thanks Jen hi everyone nice to be here today so I'm going to start by talking to you a little bit about the household finance statistics transformation project so in April this year ONS started this project and it's a three-year project but we do see it running longer than that and just to kind of give you a little bit of information about what's actually in scope it's the three surveys listed here so we have the living costs and food survey the survey on living conditions the wealth and assets survey and alternative data features in there as well so this is the aim of the HFST project and this was kind of drafted by committee hence why it's so long but I'll move on to give you the simpler picture so essentially what we're talking about is we're going to transform towards one survey where we can look at income expenditure wealth and living standards if you like for the same households over time and that that one survey is in inverted commas intentionally because at this point we don't know what the design is actually going to be it may end up being one sample and several surveys coming off it so that's just a caveat there and essentially what we want this project to deliver is more timely better quality statistical outputs we want to exploit alternative data we want the surveys underpinning the data to have greater flexibility and responsiveness and crucially we want the design and the content driven by what users really want and need so kind of big picture what it covers if you think about the outputs at the moment we have micro data from the three surveys which are used both in ONS and outside of ONS and across the three surveys there's variable timeliness and there's issues some users have in terms of granularity and other things but the data that we produce feeds into lots of important things like CPI, national accounts, tax modelling, policy analysis and lots of research the ONS publish outputs on income, wealth, spending, financial resilience and many other things using these data sources and the ONS also produces several income publications and outputs from both survey and admin sources and as you know so does DWP so in terms of the vision of the project what we want is that our household finance statistics are designed to reflect what users want to need and that we're able to give a holistic picture of financial wellbeing of the UK population alongside that we want to work towards simplifying the income data landscape so that it's more coherent and meets users needs thinking about the survey design at the moment we have three large complex surveys each with different sample designs different processing systems there's issues across the piece with sample size, historic content, timeliness and complexity of use so where we want to get to then is this sort of one survey or sample to look holistically at household finances and demographics for the same households and we want the content of that to be really user-need driven and we're hoping to move away from a kind of survey-based approach to a topic-based approach so that users of our spending data say rather than the LCF specifically we also want to be quite radical in design solutions not constrained by the current surveys and processes and we want the design to be flexible and adaptable so that we can respond to emerging needs and then alongside we're also replacing the legacy data architecture and processing systems which although is a very kind of internal ONS thing hopefully for you that means better quality data that we're able to get out to you quicker on the use of alternative data at the moment within ONS we use some income admin data in our small area income estimates and some of the adjustments on the income survey data and within ONS is a very complicated income landscape where we have several outputs produced in a mixture of admin data and survey data so in terms of the HFST vision what we want to do is review the alternative data sources for wealth and spending as well and incorporate those for quality assurance and developments as appropriate where previous focus has been on income question replacement in terms of the use of alternative data what we want to do is kind of comprehensively assess whether that's feasible and over what time frame and if it's not developed alternatives as mitigation and generally across the piece we want to use we want to make the most of alternative data to complement the survey data and use it in quality assurance and so on this is just a kind of picture of the work streams that we have on the projects sorry if you can hear my dog this just kind of shows you the breadth of the project and essentially what it's covering so we have separate work streams for things like the user needs and statistical design one for alternative data a specific work stream on coherence and how we're going to decommission the existing surveys in terms of where we're at the moment so as I said this project kicked off in April so it's still very much early days but we have already been doing extensive user engagement we've been running workshops a survey and wider engagement and I've just popped our HFST email address on here in case you want to get in touch if you haven't heard from us already and you'd like to engage that would be great we're also carrying out at the moment discovery exercises so we're looking at international and national best practice and we're establishing the as is picture across the surveys looking at how they design the processing systems and how alternative data is used at the moment a really important one is that we're also running lessons learned exercises so many of you will be aware this is not the first time that ONS has tried to transform its household finance statistics so we're learning from those previous attempts to avoid repeating the same things again and also we're working on tactical solutions alongside so we're redeveloping the processing systems in particular for the LCF at the moment and we're looking at more modern data collection tools for the living costs and food survey in terms of the next steps for the project in the short term we're going to continue work on the tactical improvements and alternative data research going to carry on engaging to establish the user needs and the business needs that the new design will be based on then we're going to explore the options for what a future design could look like and alongside we're going to have a new external governance structure so that external users are able to feed into the design decisions being made towards the end of this year we hope to have a preferred design option which we're going to put out for public consultation and then looking further forward the the second year of the project is around development and building any new systems the third year is pilots and testing and as I said it is a three-year project but we do see it continuing in the long term because as I said we want it to be kind of responsive and adaptable so as for example new alternative data comes in or there's emerging data needs in the years that follow we're hoping that we can continue this improvement and I've popped the email address there again just to kind of really prompt anybody who's interested in finding out more to get in touch with us because we'd love to hear from you I'll just stop there and hand over to Ainsley Hello everyone I'm just going to get my slides up so it's great to be back at the family finance surveys user conference and I'm Ainsley Woods and I am the coherence lead for income and earnings statistics so today I'm going to just talk a little bit more broadly about statistical coherence why and then sort of really zone in on why it's challenging for income and earnings statistics and then I'm going to talk about some of our recent work we have published an income and earnings coherence work plan we have an income and earnings interactive tool and we have updated an income and earnings statistics guide I will take you through all of that in a bit more detail I'm going to talk about OSR's office for statistics regulations review of income based poverty statistics and I've already seen that mentioned a few times by DWP which is great to see and then I'm going to cover the what's next for us so you know in terms of statistical coherence you know when we talk about that within the office we tend to use the definition that's on the screen so it's about drawing together outputs on the same topic to better explain the part of the world that they're describing you know this could be across the four nations of the UK or in the case of income statistics that can be sort of across the different producers it's worth mentioning that we do have a coherence work program at the I1S and there's a link to it if you want more detail but obviously income and earnings is part of that coherence work program so time now to zone in really on income and earnings statistics and why it's so challenging to achieve coherence in this area so first of all as hopefully you're well aware of there are three main producers working in this space there's the I1S there's the DWP and there's HMRC in terms of then the data that we can use we can get data from I'm going to say it's from three main places we can send surveys to employers asking them how much they pay their employees which is generally how we get our earnings information we can also send surveys to our employees via household surveys we can ask them how much they get from income and earnings we can from sorry earnings sorry we can also ask them about their other sources of income and then of course I suppose there's still relative new kid on the block but you know it's been there for a while is admin data but we're still very much working you know how best to use their data so it's sort of the the new one that's coming on board when we talk about income and earnings statistics they can be in terms of an individual you know we might be interested in how much an individual earns but often when we talk about income we you know users and producers you know are more interested in maybe how much a household what household income might be and then finally you know we are we work in different frameworks as well so these statistics that exist within different frameworks so we have sort of a micro framework which is very much with a household survey set but we also have the the macro or the macroeconomic framework which is basically a system of national accounts and within that system of course we have you know we have measures of income we have measures of expenditure so that's you know another layer of complexity and then this is an updated diagram that some people might be familiar with not going to go through the detail of this in any way shape or form but just you know really I think makes it really clear that there are many different stages as well to think about users are interested in outputs at different places within this sort of flow diagram so that's sort of the final kind of layer of complexity here in terms of what we're trying to do in terms of coherence work so we are really here as sort of facilitators coordinators collaborators bringing producers together getting them to you know think about the big picture and getting them to very much think about sort of cross GSS I think it's worth mentioning that we are most definitely not the output managers and we don't know them in the new show of detail that that those people know so in terms of our recent work it's worth mentioning that we have what we call our income and earnings coherence staring group that brings together senior leaders from DWP, HMRC and ONS and they sort of overseer oversee and steer a lot of the work that we do we have published an income and earnings coherence work plan and an associated blog it's very much a cross GSS work plan focused on making improvements in five key areas which I'm going to go into a little bit more detail on on the next slide we have also launched our income and earnings interactive tool which I like to describe it as a bit of a one-stop shop income and earnings statistics it's a place where users and even producers anyone interested in income and earnings statistics can go it's filterable it's searchable so you can go there if you've got something specific in mind you can use the filters in the search to find it or if you're not really short you need you can kind of help limit down the options by using the filters and things and then finally we have updated the income and earnings statistics guide and there's a screenshot of it there for you if you're familiar with this document you'll know that it is it pulls together a huge volumes of information so we've refreshed the introduction section of this updated figure one in there which is that stages of income and earnings that you've just seen we've included three new outputs that weren't published when the guide was first published we've done a lot of work to refine information in that guide especially on the strength things like strengths limitations and uses and then finally we have made it clear where we're referencing and talking about income based poverty statistics so I'm not going to spend too long on this slide just because I'm I'm conscious of time but so income and earnings coherence work plan then I talked about sort of these five key areas and these are just screenshots from the work plan which now sits on the analysis function web page so there's coherence of narrative coherence of sources accessibility quality and of course what's that's really the heart of what we're doing here is user engagement OSR's review of income based poverty statistics so as a number of people have already mentioned this was published May last year and we've been doing a lot of sort of the coordination and behind all the hard work that that producers are doing and government departments are doing to address these recommendations so a number of initiatives in our work plan already cover some of the recommendations in this report and we are currently coordinating a bit of a progress update so you've obviously seen kind of a smattering of updates through some of the slides that you've seen this morning we're sort of pulling together a bit more of a comprehensive update against this which we intend to share with OSR in the autumn and then I suppose what's next for us we have really a big part of our work especially I think for the next 12 months or so is our involvement in the household financial statistics transformation work which you've just seen Carla present about and there is now that coherence work stream which we are leading on we also have and are doing a bit of work in the space of admin data we had a workshop recently where we got users together to talk about how they're using admin data to think about how we're working together to use admin data and to think about whether there's any synergies in the work that we're doing as I've already mentioned we're doing a bit of the coordination behind that progress update for OSR against OSR's income-based poverty statistics review we are very much thinking about updating our income and earnings coherence work plan but we don't have a clear time frame for that just yet but watch this space and then finally it's worth mentioning that we did have done a lot of user engagement in the past 12 months but we are kind of bringing ourselves in under the HFST umbrella because the HFST project is just doing so much user engagement at the moment and made sense for us to kind of come in under that umbrella however we still really do welcome user feedback user engagement specifically on accessibility and coherence issues but also if you've got any feedback on any of the sort of those products that we've shown you then you know please get in touch please let us know because if we don't have that feedback then we just you know don't know if the products that we're putting out there are sort of you know suitable for our users and if there's anything that we really need to be doing to to improve those so if you do want to get in touch with us so as I've already said my name is Ainsley Woods I've also got Ian Borumum and my team sorry and we have a joint inbox for queries which is gss.income.earnings at ons.gov.uk all right I think I'm going to pass back to Carla now yeah thanks Ainsley okay so we're going to move on switch gears a little bit now to some analytical work that we've been doing let me just find my slides first of all I'm going to give you an update on some analysis that we've done using the wealth and asset survey I'm standing in for Hillary main wearing today because she's not able to be here just mainly so that you can make sure you direct any tough questions her way that'll be fine so what I'm going to talk to you about is some work we've done looking at inequalities and individual wealth and some interesting experimental work that we've done looking at the joint distribution of income consumption and wealth so we published round seven of the wealth and asset survey so that's still pre-pandemic data we published that earlier in the year and of course you know as usual there's still high levels of wealth inequality across GB households with the wealthiest 10% holding almost half of all the wealth in Great Britain and as I'm sure many of you are aware that's likely to be an underestimate we know that we need to better capture the wealthiest end of the distribution and this release of the data and associated publication was the first time we've taken a kind of in-depth look at individual wealth and the associations with characteristics of those within households so one of the things that I thought was particularly interesting was some work we did looking at regional wealth inequalities so previously we've we've only released wealth estimates down to kind of like high level regions but this time we used area classifications as well as geographic regions and on this chart here I've just I've just got a breakdown of some of the kinds of geographic regions that we're talking about so what what this classification does is essentially it puts people in the type of environment they live in rather than specifically the region so you might be you know somebody living in a rural area or ethnically diverse metropolitan living and and I think it gives a kind of really interesting additional insight so if you look in particular at say London whereas previously we might have just reported the wealth for London as a whole which always kind of comes up you know very high with this analysis we're able to look at inter regional analysis so you know as as we could see that London is a mix of very high and very well low wealth supergroups so we have the category of affluent England where medium wealth is around 227 000 but also in London you also have London cosmopolitan which has got significantly lower wealth at 37 000 so that was just kind of interesting way of making the most I suppose of the regional data that we're able to get out of the wealth and assets survey we also did some modelling so we estimated wealth disparities looking at things like age, sex, ethnic group and so on along that list and that this was really interesting as well whereas sometimes making these adjustments kind of emphasises the differences in wealth we found that in some cases adjusting for other elements actually kind of got rid of some of those disparities so the example that I've got on this chart here is if you just in the top chart this is just what the what comes out of the dataset so if you look at average total individual wealth by country of birth you can see that those born outside of the UK have a lower average wealth than those born in the UK however when we corrected for other things that that difference kind of disappeared if you like so suggesting that any difference that they have is associated with one of those other characteristics and we extended this modelling through a whole bunch of different characteristics and you know I'd really you know suggest that you go and have a look at the publication because there's lots of drop down menus where you can take a look through the modelling results and I've just put some kind of key ones on here so when we look say for example at ethnic group where the white British is the reference group you can see that there's several ethnic groups where the average wealth is significantly lower than the white British reference group that's not to say that those other ethnic groups aren't we do have some issues with sample size and you can see by the size of some of those confidence intervals that you know if we had better sample sizes that the results may well be different when you look at education you know the differences is very stark so compared to somebody with a degree qualification somebody with no qualifications is around 200,000 pounds worse off similarly with the disability with the reference person being somebody who doesn't have a disability somebody who has a disability is over 60,000 pounds worse off okay now we're going to move to an interesting experimental project that we did so I think it's fair to say that with my HFST hat on as well in some ways the holy grail is having spending income and wealth for the same households and at the moment you know that data doesn't exist so what we did was we statistically matched the wealth and asset survey and the living costs and food survey to essentially give us a proxy of that data set so major caveat is that these are not the same household the same households they are similar households so essentially we attached the spending of a similar household onto the wealth and income data from the wealth and asset survey and this allowed us to do some really interesting things so being able to look at somebody's wealth alongside their income and spending can you know give a bit of additional insight so in this chart here in the kind of dark blue we've got people who are spending more than their income but have got a buffer to sustain that but that buffer lasts for less than a year in light blue if you these ones expect they spend more than their income but they can sustain that for more than a year with their financial assets and then in the sort of mid blue you have income greater than spending and what we found is that working aged adults living alone and lone parents lacked financial buffers to sustain their over spend retirees who spent more than their income had a much greater buffer to fall back on as you'd expect those retirees have been able to build up some financial assets but a retired person living alone could only sustain that over spend for an average of three years whereas a retired couple could sustain that over spend for seven years in comparison we also use this combined data set to put together some poverty indicators so just to kind of let you know what these indicators are specifically so we put you as in income poverty if you have less than 60% of median income the same for expenditure 60% of median expenditure and in financial wealth you'd be in financial wealth poverty if you have a quarter of the income of the poverty threshold in financial assets so really interesting that sort of a lot of households were in poverty across all three so estimate around two million households and financial wealth poverty was you know the most prevalent so that sort of ties into the work on financial resilience where you know households don't tend to have much in terms of liquid financial assets to cover things like reduction in income which is particularly important obviously for both the pandemic and the cost of living crisis we also had a look at this by region so around half of households in the north of England were in financial wealth poverty and the northeast also had the highest levels of income poverty London and the west midlands were most likely to be in spending poverty and the west midlands is most likely to be in poverty for all three measures followed by London Yorkshire and the Humber and the northeast so I hope that's given you a little bit of a flavour of some of the analytical work that we've been doing in the kind of wealth and pensions and spending space and I'll hand over now to Ann Marie who's going to talk to us about household finances and the cost of living Super, thanks Carla Can you hear me all right? Yeah Great, so I'm Ann Marie De Silva and I'm a senior researcher in the Income and Wealth Division at the ONS and I work on cross cutting analysis for household finances so my presentation today will be taking you through some of the analysis that the ONS has done looking at the cost of the changes in cost of living over the past year so I'm going to begin by just taking you through a couple of the options of surveys that we've been using in order to do this analysis the bulk of my presentation will be taking everyone through some of the findings that we already have and I'll end by kind of sign posting some upcoming work that we have so on the slide you can see some of the surveys that have been used to look at the cost of living the first survey that's on there is the Opinions and Lifestyles Survey so that's one of our more rapid rapid turnaround surveys it is a fortnightly survey that's administered with about 2000 to 2500 households across Great Britain and there is a fortnightly analysis that's published on our website called the public opinion and social trends so you can take a look at that for ongoing insight into how people are faring not just on household finances but on several other contemporary issues the rest of my presentation will largely look at findings from the OPN the other two surveys the living costs and food survey and the household finances survey provide insight on people's spending and income respectively but these are annual publications we're going to be having publications using these surveys in the next couple of months so and I'll talk about that at the end of the presentation so what we've been seeing since data collection in November 2021 is a steady rise in adults reporting a rise in the cost of living so in November 2021 we saw about six in 10 people stating that they were experiencing a rising cost of living and by March 2022 that number had risen to nine in 10 and for the most recent data collection in June the figures still stayed at nine in 10 adults reporting this rise in cost of living the main reasons reported for high cost of living has been quite consistent with the top three reasons being an in order an increase in the price of a food shop increase in price of gas and electricity bills and an increase in the price of fuel people are generally finding themselves in difficulty meeting household regular household bills so in particular energy energy bills there's quite a few people saying that they they're experiencing some some amount of difficulty so in this reference period in March there were 43% of people so say four in 10 people reporting that they experienced difficulty in in more recent periods as well that figure has been about five in 10 somewhat stabilizing and in March there were six percent of all people reporting that they're behind on gas and electricity bills so notably this is just before the energy price cap changed on the 1st of April the number of people reporting being behind on gas and electricity bills has has remained somewhat constant since that time though we're also seeing that renters have been more find themselves in relatively more difficult circumstances than than comparative groups so there's 13% of renters who report being behind on bills compared to only 3% of people with mortgages and compared to 2% of people who own the house that they occupy 34% of renters reported that their rent had also increased in the in the previous six months from the reference period this also coincides with the same period that showed that rental prices in the UK had gone up the most since 2016 so this is this is March 2022 compared to 2016 and yeah so there's there's a lot more renters that are reporting this kind of difficulty in in affording housing costs thanks next slide please there's also some insight that we have about how people so in addition to what kind of difficulty people are facing in paying in paying bills we also have a measure of financial resilience so there is a question about how whether or not somebody can afford to pay an unexpected but necessary cost of expense of 850 pounds what we've seen since November is that the the percentage of people who are saying that they can't afford this sudden expense has been stable since since November but I think what's interesting is when you dive down into the demographics and it's evident that there are people from different demographics that are more financially vulnerable than others so in this chart you can see one of the one of these main characteristics somewhat unsurprisingly people on lower incomes more frequently report that they're unable to to short this kind of unexpected expense but what is interesting I think is the is the odds ratio so if you look at anyone earning below 20,000 pounds per annum they are between eight and nine times more likely to report that they're unable to shoulder that cost compared to somebody earning above 50,000 pounds per annum and then even if you're earning between 20,000 and 30,000 pounds you are up to six times more likely to be unable to afford that kind of sudden expense there's other characteristics that were studied as well the second the second kind of characteristic that showed more likelihood of being unable to afford this kind of expense were people who were renting so there were 53% of renters who stated that they wouldn't be able to afford this sudden expense compared to only 13% of people who own the house that they occupied and renters also were six times more likely to be able to say that they wouldn't be able to afford this kind of expense compared to only two percent so two times likelihood for people with mortgages and this is all in comparison to people who own their homes thanks next slide Carla we've got five minutes including questions okay I'll speed up then so how are people managing this rise the the main way that they kind of navigate this is to spend less on non-essentials but they are also doing things like shopping around more to find cheaper products they're using less gas and electricity but that is also somewhat seasonal because we're moving from winter into summer and they're also spending less there's to a lesser extent they're spending less on food shopping and essentials so since January 2022 we've seen an increase in the the number of people saying that they are spending more in order to buy the same amount of goods and we're also seeing concurrently seeing an increase in the number of people saying that they're buying less when they go food shopping in terms of management methods something notable is that we haven't seen a significant change in people using credit or borrowing money in order to to manage costs and we think that this might this could be attributed to to savings built up over the pandemic due to force cuts in expenditure next slide please and of course this has an impact on people's levels of worry in by May 2022 we were seeing that three and four adults reported some level of worry about the rising cost of living and 80% of those people were were thinking about were worried about it for several days or more in the two weeks prior to being interviewed although there's generally similar levels of worry across Great Britain some demographics of people are slightly more report being more worried than than others so women are more likely than men people who are aged between 30 to 69 and more report more frequent worries than those aged above 70 disabled people are more worried than non-disabled people and parents with dependent children under four compared to parents without dependent children but these differences are around 10 to 15 percentage points of each other so that's a glance into some of the insights that we're able to that that we currently have at the ONS but we also have some upcoming work as well to keep on the horizon so there are two publications coming out in summer there's a family spending in the UK bulletin that'll be out July 18th and this looks at expenditure household expenditures in the financial year 2021 so that covers spending during the pandemic so it'll provide a good snapshot of where people's finances are just prior to to the change in cost of living and we will also have the publication on the effects of taxes and benefits on income which essentially is what it says on the tin just looking at how indirect and direct taxes affect taxes and benefits affect income and we're also looking into some experimental statistics using the household finances survey so like I said household finances we we have annual estimates drawn from that but we're trying to look into creating faster more timely insights by creating quarterly estimates from there so we're testing the feasibility of creating quarterly estimates at the moment thanks