Income Insecurity and the Relational Coping Strategies of Low-Income Households in the UK

Abstract Changes in the labour market, high rates of working age poverty, major welfare reforms and more recently the Covid-19 pandemic have drawn renewed attention to income security. Existing research has identified the important role of relational support in helping people cope with low income, but less is known about the role of support for those coping with the potentially destabilising effects of income change which can affect people over relatively short periods of time. This article focuses on how relational coping strategies are utilised by those experiencing such income change. The data are drawn from a qualitative longitudinal study of the experience of income change and insecurity in 15 low-income households in the UK which included repeated in-depth interviews and weekly financial diaries completed in periods of up to five months. The article explores the relational strategies adopted by participants to ‘get by’ as well as examining how strategies are adopted by those on different levels of low income and with differing networks. The article argues that these strategies illuminate the importance of income change in the experience of low-income households, develop the concept of income insecurity, and provide lessons for policy in providing flexible and responsive support when income changes.


Introduction
Since the financial crash of / and its economic and social consequences, income insecurity has received increased attentionwhich has intensified during the Covid- Pandemic. Changes within the labour market and social security systemparticularly, in the last two decadeshave posed new challenges to the income security of working-age low-income households. Within the labour market, there have been a number of drivers of income insecurity. Before the Covid- pandemic, the UK recorded the lowest rates of unemployment since December  (.%) and the highest levels of employment (.%) since comparable records began in  (ONS, b). But these high rates of employment were in part due to an increase in part-time work, self-employment and non-standard work contracts (Taylor, ). Alongside these developments in the labour market, there have been substantial financial cuts to the social security budget and an ongoing implementation of radical reform to the UK social security system (Millar and Sainsbury, ). The central feature of this reform has been the introduction of Universal Creditwhich replaces six benefits and tax credits  with a single means-tested system, available to people both in and out of work (Millar and Bennett, ). Brewer et al. () estimate gains and losses under Universal Credit and conclude that those with temporary falls in income are amongst those who lose out (p.).
The combination of low and unstable income seems likely to have a different impact on living standards from living on a low but steady income (Morduch and Schneider, ). It adds an additional challenge of coping with changes, as income goes up and down, possibly over quite short periods of time. And, with the possible reduction in traditional forms of security from the labour market and the social security system, a greater role may be expected of family and friends.
This article will focus on the role of relational support, defined as financial and non-financial support provided by family, friends and acquaintances, in coping with income change in short time periods. It is based on an in-depth longitudinal study of  means-tested benefit claimant households in the UK. The research focuses on income change over relatively short periods (less than five months) and how this affects or interacts with coping strategies. There are four main sections: the first examines the literature on income insecurity and relational coping strategies; the second sets out the research methods of the underlying research; the third presents the findings of the wider research on the role of relational networks in how low-income households cope with income change. The final section discusses the implications of this research for the conceptual development of income insecurity, and for policy.

Income insecurity and relational coping strategies
Insecurity is a widely discussed policy issue within the UK (Citizens Advice, ; Shafique, ) but less attention has been paid to its conceptual development within social policy literature (Orton, ). While there are many different forms of insecurity that are central to the experience of low-income householdssuch as economic, housing and food insecurities (Eurofound, ; Garratt, ; Hardie, )there is a need for a more specific conceptualisation of income insecurity. Income insecurity involves a lack of income adequacy and predictability over time as well as changes to "actual, perceived and expected income" (ILO, ). This definition stresses the importance of both income change (whether income goes up or down) and income variation (the difference between recorded income over time and the average for a longer period).
The dynamic nature of income receipt poses challenges for the way in which it is studied and measured. One-off surveys, for example, can fail to capture income dynamics from one year to the next (Jenkins, ). Further, largescale surveys can miss income change within the year altogetheraggregating participants' income and leaving policy makers to assume average monthly income receipt. Income change within the year has received limited attention within social policy related literature. Hills et al. () deployed two-weekly income diaries to look within the year for a sample of families in the UK. They found income variation differed by source of income and was higher for those on the lowest incomes. Two more recent studies, one based in the UK and one in the US, use financial diaries to focus on month-to-month financial volatility or the degree of variation between actual recorded income and expenditure and that received over a longer period. Biosca et al. () found a large number of financial transactions and that, on average, participants were making important financial decisions every two days in order to manage shortterm financial changes. Morduch and Schneider's () US study found that this income and expenditure 'unsteadiness' in short periods was a defining part of the experience of low income households in their sample. Monthly income volatility has received recent policy research interest too. Tomlinson () looked specifically at wage variation in the UK using Lloyds banking data from seven million UK accounts, and found pay fluctuations were the norm for the majority of employees in their sample. A common finding amongst these studies was that income changes in short periods were more common for those on the lowest incomes. This body of literature provides important insights into the incidence of income change but few studies have considered the experience of income change in short periods.
How households cope with low income relies heavily on the resources they have available over time. Getting by is defined by Lister () as a form of coping that is often invisible or ignored until a problem is identified. The ability to get bywhich could mean, for example, finding ways to obtain food, access childcare or reduce stressis unequally distributed amongst people and can be constrained by a lack of access to "personal, social and material resources" (p.). Although the term 'strategies' is widely used to describe this form of agency, this concept is better understood as having explicit and implicit elements (Dewilde, ). While the conscious weighing up of options and choosing one over the other is important, so is a more implicit " : : : adherence to an existing course of action" (p.). A focus on the act of 'getting by' also allows for a better understanding of the experience of income change in short periods. This process of getting by often involves relational networks, here defined as family, friends and acquaintances that might provide monetary and non-monetary support, help and encouragement. When labour markets and social security systems fail to provide income security, familial, friendship and wider networks can play a crucial but sometimes limited role (Daly and Kelly, ; Edin and Lein, ; Rebollo et al., ). Important limiting factors include stigma, reciprocal obligations and the limited resources of relational networks (Hannon, ; Hill et al., ; Patrick, ).
There is some evidence of the role networks play in the lives of those on a low income, as a source of financial help, support with childcare, housing and food, and emotional support and advice (Hill et al., ; Millar, ). But less is known about the importance of financial help at the right time. The financial stability of those providing the support is also crucial as Morduch and Schneider () show in their work with low to middle income households in the US. They found people relying on the stability of those in their networks when managing their own income volatility. Thus, availability and timing of support are both crucial.
Network support can also have associated complications and costs. Reciprocal obligations complicate many relationships with expectations of a responding contribution. Edin and Lein () found that alongside smoothing income came the role of "equalizing distribution within the network" (p,). This expectation can have social and practical consequences, damaging relationships when contribution is unequal and stalling movement out of poverty for those who have improved their circumstances and find themselves compelled to help others in their network (Hannon , p.). There can also be negative consequences for those who help and who experience further hardship themselves (Cohen et al., ).
Income insecurity is not just about low income but about income change. It is therefore important to look in more detail at the extent and nature of income change and how people cope with that in their everyday lives. We know that family and wider networks are important in managing low income and the aim of this analysis is to explore in more detail the relationship between income change and relational coping strategies. These strategies are complex socially embedded phenomenon that have been explored in other research but often without a conception of change over time. How a household copes one day may not be how they coped last week or the month before and a key element of this is income change. People on low incomes live in real time and with increasingly unstable incomes; so a focus on income change alongside relational coping strategies gives added insight into how they work together in people's lives.

Researching income change and insecurity
This research took place in a city in the UK between December  and February  using a qualitative longitudinal approach in order to reveal complex lived experience over time (Corden and Millar, ; Saldana, ).
Qualitative longitudinal research was thought to be best placed, over relatively short periods of time, to investigate change that is often excluded by larger scale snapshot research and to provide real world/time insights that can inform policy (Neale, ). The main aim of the study was to explore how low-income households experience income change and insecurity in short periods. As discussed above, here we focus on one aspect of that -the role played by relational networks in how respondents cope with income change.
The sample criteria were to include people claiming: at least one meanstested benefit (and therefore on a relatively low income and a member of a 'household' for benefit purposes); of working age; and with good knowledge of their household finances. A mixed sample of  households was recruited through different local organizations and flyer handouts. A third of participants were clients of one advice organisation and the other two thirds came from organisations including a housing association, a children's centre and a carers support service. A decision was made not to recruit from a single advice organisation to avoid a sample of participants with pre-existing problems.
All in the sample were interviewed at least once but, in order to focus on change over time, those who agreed to further interviews carried on with the research. Nine were interviewed more than once and five completed income and expenditure diaries over periods of up to five months. Those who carried on did so for a variety of reasons, including having the time and inclination to continue but no decision was made to encourage particular people. All participants faced challenges that made participation a commitment and this was addressed through relationship building, incentive vouchers of £ per meeting and flexible participation requirements. However, it must be acknowledged that the voices of those who could not continue are less present.
In each of the  households one adult was interviewed about their household. Ethical approval was obtained through the University ethics committee and data was anonymized and securely stored. The participants provided informed consent and have been given pseudonyms to preserve anonymity. Data was organised and analysed using NVivo-, a computer assisted qualitative data analysis software package and framework analysis methods (Ritchie and Spencer, ). The sample of  households included five single people, six lone parents and four couples with children, ranging in age from  to . Most () rented from a housing association with two in the private sector and two in owner-occupation. There were ten households including someone with a disability. Ten households had at least one person in work (including three self-employed) and twelve were receiving Universal Credit with three claiming other means-tested benefits. Most participants () also reported income change from one payment to the next. Three had experienced change in just their work income, five in just their benefit income, four in both their work and benefit income, and one in work and other income. The diary data (for five households) allowed for more detailed real-time tracking of these changes and are drawn on in the analysis below.
Eligibility for means-tested benefits was as a proxy for low income within this study and household incomes varied between £ and £, per annum. Household circumstances make comparison difficult, but this meant some households had relatively low or high incomes. Figure  shows reported monthly income at the first interview by household type. This variation was consistent with relative conceptions of low income and useful in what it could reveal about the different relational coping strategies adopted. For some, income dips in short periods meant a drop in living standards, while for others they meant severe hardship. Individual household circumstances will be set out below, including details of household's ability to save, current debt and foodbank use.

The role of relational networks
Participants relied heavily on relational networks to get through periods of low income. During the analytical process, the sample was divided between one group who had family and friends with the security and willingness to lend money and support when needed, a second group who sometimes had support when needed, and a third group who struggled to borrow or get support at all. This section will look at the role of borrowing and support for each of these three groups, giving detailed examples of short-term income change and how relational strategies were adopted to respond.

Help available when needed
First, were those who could rely on the right support at the right time. This could involve borrowing money to smooth income as well as non-monetary support to address need. Most participants (/) in this group had the highest FIGURE . Monthly income by household type Source: Interview data incomes in the sample, multiple income sources and one or more children. In their first interviews, four of six participants said they felt able to save, one was in debt and four had recently used a foodbank. Lending and support were often delivered in a flexible way by those who had an understanding of participants' situations. This could have complications, but participants generally received relational support in times of need.
The Bennetts were an example of this flexible relational lending and support, available when needed as income changed in short periods. Sue and Rob Bennett lived in social housing with their two children aged  and four months. In their research period, Sue was on maternity leave from an administrative job and experienced changes in her monthly maternity pay and Rob went from working as a manager at a retail outlet with four-weekly pay and unpaid overtime to working as a weekly paid driver. The Bennetts' estimated monthly income at the time of their first interview was just below £, which was relatively high within the sample but below the average for the UK of £, per month (ONS, a). They were experiencing income variation but felt able to cope because of the willingness of Sue's mother to help when they needed it. Figure  shows the Bennetts' weekly income variation when their four sources of income are combined and set against their average income for their fifteen-week research period (orange dotted line) and their estimated yearly income (grey dashed line). It shows a regular pattern of household income until week seven, when their income drops as Rob moved into a period between jobs and then started his new weekly paid role.
During the study period, the Bennetts identified two periods of financial difficulty following the birth of their second child: waiting for their first Universal Credit payment between November and February (week  of their study period) and the dip in their income as Sue's maternity pay went down and Rob changed jobs, between week  and week . Rather than being temporary, these periods were seen by Sue Bennett as part of ongoing income change: "We were getting no help for quite a while, so then we needed the [Universal Credit] payment in February and then that was less because of this charge [deduction] so it's just kind of trying to juggle money in a different way now, and then with him changing his job, that's going to be another change and I got paid about £ less than I usually would. It's like lots of different changes all in, at different times, you don't have enough time to adjust to how your payments are because we've had so many different things." The Bennetts were not in debt and could sometimes save, but short periods of low income posed challenges. Benefit payments also did not come in the right amount, at the right time to address income inadequacy and they identify Sue's mother as the provider of financial support when they need it.
In the first period, which lasted for  weeks and included Christmas, the family were left without the extra money they were entitled to from Universal Credit. This was because of a mixture of the built-in waiting period (then six weeks) and a monthly assessment that counted two of her pay packets as a full month's pay. In Sue's words: "We have essentially gone from October to now, what will be February, until we get any financial help with our income when actually we should have already had some kind of help and there's no fall back, there's no help for you in the meantime, so I'm lucky I'm one of the fortunate people, I have my mum who can help me" In this period, her mother regularly paid bills and bought food, clothes and toys for the children. Rather than one off support, Sue described her mother's help as ongoing but increasing at times of need. This financial support intensified during the second period of low-income they identify when Rob changed job, Sue was being paid less and Universal Credit had not yet caught up.
Sue describes a relational and needs based arrangement that differs from formal borrowing such as pay day loans in that the amount borrowed and paid back was based on a shared understanding of her need, what her mother could afford and what she could repay: "I'm just really lucky that my mum has the income to do it, not everyone is in the position that I'm in and has a mum that can just step in whenever I need money : : : as long as I pay her back over time, it's a lot easier than thinking I have a huge bill." While there was always a future intention to pay her mother back, financial needs such as those in the period where she wasn't receiving her Universal Credit were sometimes just covered: "I didn't borrow she just gave me some money". The support of her mother represented an important form of security in her life, cutting expenses, smoothing income through periods of change and countering feelings of not knowing where support would come from. Her mother's help amounts to a continued form of financial support with a built in understanding of their circumstances.
The responsive network support available to the Bennetts was common within this group with closer family members such as siblings, parents, and friends providing financial, in kind and emotional support when needed. Within the wider sample, five other households could be categorised as having help available when needed. These households tended to be those on higher incomes with multiple sources of income and one or more children.
An exception within this group of households was Richard Jenkins who was single, had the lowest monthly income in the sample at the time of his first interview (£), and was experiencing severe hardship. He had no savings, was in debt and regularly used the foodbank. He was also unable to work because of multiple health conditions, some of which were variable in the effect they had on him; he claimed income-related Employment and Support Allowance (ESA), a means-tested income replacement benefit, and Personal Independence Payment (PIP), a non-means-tested disability costs benefit. His ESA was stopped after an assessment of his work capability and he was left to live on £ per month. Richard identified a pattern of failing his regular disability reassessments, which were then re-instated after appeal: He was currently experiencing a period of low income and relying heavily on neighbours. In particular, he identified good friends within neighbouring flats, with whom he shared favours and money as well as relying on the security of their credit rating and pay periods. One close friend would lend him money when she was paid and expect it back when he was paid: "I've got a friend that gets paid next Wednesday, and she usually lends me a bit of money and then I pay her back when I get my PIPs on the Thursday following." This strategy, which was adopted by others in the sample, shows the importance of timing in this form of support. While Richard had close friends able to plan their lending according to payment periods, other households identified the difficulty of not having money at the right time as well as not having the money at all. For Richard, his income inadequacy was addressed over time by the combination of a friend willing to borrow money on her credit card, for him to pay back over time, and a friend who lent him money according to their respective pay dates. This strategy involved the mutual trust that payments were reliable and could smooth income as it went up and down, with participants effectively providing informal financial services to each other. For Richard Jenkins and the Bennetts, flexible solutions to income change were reliant on relational understandings of need and the ability to meet it over time. While their incomes were different, both show how relational support is used over time to smooth income and cope.

Help sometimes available when needed
Second, were those with relational networks that were less willing or able to help at the right time. These households tended to be on lower incomes and have fewer sources of income than the first group and all had children. They were also more likely to borrow from different sources and face higher costs and consequences when receiving money and support. In their first interview, one of four participants said they felt able to save, three were in debt and all had recently used a foodbank. Relational strategies again included flexible support but also last resorts when this was not available.
The Lennon household were an example of the central importance of relationship quality. Samantha and Paul lived in social housing with their one-yearold child. In their research period, Samantha cared for their child, and her partner Paul was on a zero hours contract  with fluctuating hours of work. They were in a significant amount of debt and had previously borrowed from friends and family but these borrowing options had been restricted by not being able to pay money back. They also regularly used a foodbank and were unable to save throughout their research period. Samantha also said she missed meals and withheld rent in order to save money in low-income weeks.
Their estimated monthly employment income at the time of their first interview was £, which was just above the average income within the sample. They experienced a period of high income because of overtime and then low income because of illness at the end of the research period (see figure ). Samantha reported feeling very insecure about their situation and that help had sometimes been available when they needed it in the past but that this had relied heavily on the discretion of the giver. Figure  shows the Lennons' weekly income variation when their three sources of income are combined and shown against their average income for their fourteen-week research period (orange dotted line) and their estimated yearly income (grey dashed line). It shows a relatively stable weekly work income and monthly benefit income for ten weeks before a period of Christmas overtime increases it significantly to £ in week  and then a period of sickness absence decreases it to £ in week . Paul's income also interacts with their Universal Credit, which goes from around £ in week  to £ in week  because of an increase in income in the previous month.
Although, as figure  shows, Paul's wages can be quite consistent, he does not have guaranteed hours and loses money when he is ill. Debt was a crucial factor in the Lennon household and meant that spikes in income such as that in week  of the research period were often used to pay off debts and so did not greatly improve their financial situation. In week , when Paul's income spiked to £ because of a Christmas overtime lump sum payment, the household faced an increase in outgoings, which included £ towards petrol and buying a new car that was essential for Paul's work, £ on increased rent payments including arrears, and £ on increased Christmas food spending. They also paid £ in bank charges on Paul's overdraft and money towards two outstanding pay-day loans. Multiple debts and the consequences of not paying them left them feeling out of control: "We're getting more and more into debt, we had a pay day loan that was meant to come out last Friday, he didn't pay it, we didn't have the means to pay it, so now that's going to be more and more and more, bank charges,  quid a month, it's everything spiralling out of control." Periods when Paul's income was stable were already difficult because of the payment of debts but periods of lower income required further borrowing. Their Universal Credit also dipped at the same time as Paul's income deepening their income fall and further limiting the adequacy of their income. From a policy perspective, Universal Credit is responding to a big increase in income in a claimant's monthly assessment period. However, debt constrains the Lennons' ability to save when their income increases, and they must deal with low-income months such as week  when Paul is ill and receives £. The compounding and spiralling impact of debt can also be seen here, in that one of the Lennon's strategies in week  was not to pay their rent, causing rent arrears that would increase their future bills. Not being able to pay back previous loans shaped Samantha's relationship with her family and restricted what they felt able to ask for when they needed it: "I don't want to be a charity case, when we're family it's very contentious when it comes to things like money and support, as I owe them a lot of money even though they've written it off, I don't feel like I can phone them up and say I'm struggling to pay the rent this week." This sense of not feeling in control of whether and when they get support created what they identified as a sense of insecurity. Samantha also described an assertion of control and influence over their lives from those to whom they owed moneyin particular, from closer family. She felt that money and support came with conditionsfor example, how they were expected to behave and what they bought. This meant borrowing from close family and friends was not always possible and they had to look to wider networks, often as a last resort.
When thinking about replacing their old car which was essential for Paul to get to work, they identified two main borrowing options. One was Paul's boss who offered to loan them the money at a high interest rate and then withdrew the offer because he thought the car was unsuitable. The second was an informal payment arrangement with the owner of the car: "The man who's got the car, he's willing to take payments so we pay them directly into his account but unfortunately we just haven't had the means to pay him the last three weeks." They had already paid £ to this person but had no receipts and would not get the car until they paid the full amount. This situation was ongoing throughout the research period and was a 'last resort' saving strategy adopted after others were exhausted.
The experience of the Lennon household shows the importance of relationship quality when in need of financial and non-financial support. In particular, the feeling of being able to assert some control over when support is provided rather than waiting for others to decide to give. Within the wider sample, three other households could be categorised as sometimes having help available when needed. Unlike the first group, support was more likely to include costs and complications, to come from different sources and to involve reciprocal expectations.
Responsive and timely support was again vital to this group despite being less available and there were examples of strategies that relied on wider networks. Katy Deacon was a single parent who lived with her two young children. She had high levels of debt and a long-term mental health condition that she identified as constraining her personal and financial wellbeing. She did not have savings, had recently used a foodbank, turned the heating off when her children were at school and missed meals to cut down on expenses. While explaining how she got through a period of low income waiting for her first Universal Credit payment, Katy identified several coping strategies that involved support from her family and friends, sharing resources such as internet connections, shopping for each other and reciprocal care arrangements. Katy also makes a distinction between borrowing from a company and borrowing from a person with whom you have a close personal relationship: "I felt a lot worse for not being able to give my mum back the money I'd borrowed off her than I did because I couldn't pay my Sky bill, that's a company." This awareness of the welfare and needs of the giver was shared throughout the sample and underpinned the difficulties of relying on those who were struggling themselves. This usually meant widening the search for support beyond immediate family members. Katy also identifies a period of co-residence in which: "Two of my friends from Romania had come to visit for three months and they bought all the food in the house as their contribution to me letting them stay, they done all the cooking and the cleaning and then that meant that I could then try and get back on top of myself." Katy therefore addressed her financial and health needs through accessing a wider network of support that became available at the right time. She was also very aware of the reciprocal nature of this support.
These examples show the central role of relationships in addressing periods of income inadequacy over time and that these options can be stretched when close family members are unable to help. Wider networks were then accessed but these tended to involve issues of control, less understanding of need and clearer reciprocal obligations.

Help rarely available when needed
Third, were those who were rarely able to rely on relational networks to help them when they needed it. These households were on the lowest incomes, usually had fewer sources of income and tended to rely more on informal arrangements, charitable support networks and other available resources. In their first interview, one of five participants said they were able to save, all were in debt and all but one had recently used a foodbank. Support came where and when it was available and sometimes no support would be available at all, challenging the very idea of coping.
Rachel Martin was an example of the need to rely on wider and less reliable networks when support was not available from family and friends. Instead, she relied on charitable organisations, and a network of friends and acquaintances with whom she shared small amounts of money, favours and encouragement. Rachel's estimated monthly income at the time of her first interview was £she did not feel able to save, was in several thousand pounds worth of debt and regularly used the foodbank. She had no family in the UK and lived alone in social housing. In her research period, she worked as a cleaner for three separate companies but only one of those became permanent. Rachel had also just experienced a period of low income when she was not paid Universal Credit for two months after a cleaning job ended: this had also put her further into debt. Figure  shows Rachel Martin's weekly income variation when her work and benefit income is combined and shown against her average income for her sixteen-week research period (orange dotted line) and her estimated yearly income (grey dashed line). It shows her Universal Credit and work income in week  and her three weeks income from another job in week . This income is then paid two-weekly at the same rate in weeks ,  and  before dropping down in week  because of a decrease in hours over a holiday period. Income then rises in week  because of an increase in hours before returning to the level of week  because of the giving up of extra hours.
Rachel identified the nine-week period just before the research started as the most difficult. Because under Universal Credit, income is assessed monthly in arrears with reference to what was earned in a monthly period, her latest pay date meant that instead of getting her full entitlement as a single unemployed person she was not entitled to Universal Credit payments for two months: "The last wage I received was on  th August and then I thought I was going to get money from Universal Credit, but they said that if I got paid on  nd August, I would have got Universal Credit but since I was paid on  th I wasn't entitled to any" Because her assessment period ended just before her payment in August, she was not entitled to a payment in September and had to wait until  th October before receiving a Universal Credit payment  . Rachel was experiencing an ongoing problem of income inadequacy and change confounded by debt. In the absence of family support, Rachel coped through charitable support and a network of friends and acquaintances.
"I've got various friends where we do each other favours, let's say somebody bought me a coffee and then I gave him some crossword puzzles, things like that yeah, sort of exchange of services like you do with friends." Rachel describes a less reliable form of support, from multiple individuals who offer friendship and encouragement and low levels of monetary support. This support is rarely at the right time and is provided by people with similarly limited resources. It also involves reciprocal obligations, the shame of not wanting to be known for taking, and the recognition of need, such as a time she borrowed £ from a friend in her community: "I want to keep a friendship, I don't want to be seen as the, like a handout, and I've known him for years but no that would be mutual trust like ok then I'll buy you a coffee some other day, but the fact that he realised how I wouldn't ask for it, that he realised that I really needed it." These forms of support were not enough to avoid hardship but mattered to Rachel in the absence of family support: "It's vital, without them I probably would have killed myself, maybe". In the specific period of very low income while waiting for her first Universal Credit payment, Rachel says she was unable to access support from friends and she cut back by turning her electricity off and relying on the foodbank.
This was an experience shared by other participants in this group who had little contact with family members able and willing to help them. Support networks were thinly spread, and less reliable, charitable support was restricted and sometimes unavailable and both lacked the flexibility and understanding of need that was more common in the support provided to those in the other two groups.

Addressing income insecurity
These findings show that income change in short periods is an important aspect of people's lives and for those with low income it can create insecurity. This change adds an extra dynamic, which is under explored or ignored in previous research. Income level matters but so does whether income is received when it is needed and whether a household has the means to smooth income through relational support. Households within this study typically had low incomes, which often tended to change up and down, even in short time periods. For some, this income change could be mitigated by their relational networks and support being available when needed. Relational strategies were crucial for those experiencing income change and insecurity. For those most in need, support often came from others in need, restricting what was available and when and involving costs and times when support was not available at all.
The three groups identified above show the central importance of relationships and timing of support in coping with income change in short periods. They also show that the ability to access timely relational support may be unevenly spread amongst low-income households. The first group had support available when they needed it. This group tended to have higher incomes and experienced closer relationships with those giving money who had a developed understanding of their needs. Support was responsive and flexible, usually from the same person or people and crucially was received at the right time. The second group sometimes had support when they needed it. This group tended to have incomes that were close to average within the sample and were more likely to experience higher relational costs to receiving support. This support was less regular than for the first group and more likely to be one off and from more than one source. When closer friends and family were unable to help, support was received from wider networks with less understanding of the recipients' needs and clearer dynamics of control and reciprocity. The third group rarely had support when they needed it. This group tended to have lower incomes and weaker relationships with those giving support and may have used up help in the pastfor example, by not fulfilling reciprocal obligations. The support that was available was almost always inflexible, short-term and one off and its timing was unpredictable. When support was not available, this group either accessed charitable support that lacked flexibility and understanding of their needs or they did not access support at all and faced extreme hardship.
Within the wider study (Young, ), income insecurity is conceptualised as having three inter-related elements. These are income instability, income inadequacy, and the constraint of these on the lives of individuals and their households. This article focuses on the instability element with an examination of the experience of income change, but also the short periods of income inadequacy that it can create. As these findings show, change and adequacy are often tightly bound together and create constraints on the lives of low-income households that they address though relational support. Support that is available, responsive and flexible (and that is provided by those with an understanding of recipient's needs) is vital when income changes.

Conclusion
These findings also raise the question of how income security can be ensured through policy. Universal Credit combines benefits in and out of work and thus seeks to provide an ongoing income over time, changing as people experience changes in their circumstances, but aspects of its design and delivery have meant increased instability for some. As others have pointed out, the timing of income assessment is also crucial (Brewer et al., ; Millar and Whiteford, ). The monthly assessment within Universal Credit may mean claimants receive different amounts each month and so challenges the stability of claimant incomes (Young, ).
Balancing the need for both stability of income and responsiveness to change is a clear policy challenge for means-tested benefits. Universal Credit could better deal with income change in short periods and smooth income by providing a fixed income for a longer period. However, this may come at the expense of a degree of responsiveness. Another option would be to look in closer detail at the financial circumstances of claimants, perhaps working in partnership with support providers to understand how debt constrains claimants' finances on apparently 'good' weeks.
It may be that the conflict between income instability and responsiveness cannot be addressed within Universal Credit alone but involves wider social security provision. For example, for families with children the need for a secure longer-term income could be addressed through increasing universal benefits such as Child Benefit. However, there is also a need to address the needs of single people without children, who this research identifies as least able to access relational support. This may involve a reconsideration of the universalism of social security provision for all groups, with responsiveness to need addressed through means-tested provision that is more flexible and responsive to claimants' shortterm needs. Such a combination could better provide adequate and timely financial support to those who need it.
 Within this two-month period Rachel received an advance payment of £ from DWP which would then be taken from future Universal Credit payments at the rate of £ a month until it was paid off.