Research and analysis

Are household formation decisions and living together fraud and error affected by the Living Together as a Married Couple policy? An evidence review

Published 6 July 2023

DWP research report no. 1031

A report of research carried out by RAND Europe on behalf of the Department for Work and Pensions.

Crown copyright 2022.

You may re-use this information (not including logos) free of charge in any format or medium, under the terms of the Open Government Licence. To view this licence, visit http://www.nationalarchives.gov.uk/doc/open-government-licence/ or write to the

Information Policy Team
The National Archives
Kew
London
TW9 4DU

Email psi@nationalarchives.gov.uk

This document/publication is also available on our website at GOV.UK

If you would like to know more about DWP research, email socialresearch@dwp.gov.uk

First published July 2023.

ISBN: 978-1-78659-532-4

Views expressed in this report are not necessarily those of the Department for Work and Pensions or any other government department.

Executive summary

This report summarises evidence on living together fraud and error (LTFE) and the social and economic context in which LTFE is embedded, with a view to informing DWP’s living together policy and its application to reserved GB benefits. LTFE refers to situations where a claimant has a partner but is receiving benefits as a single person (whether intentionally – i.e. fraud – or due to error). LTFE is a widespread issue in the UK, costing the taxpayer an estimated £760m per year.^{[footnote 1]} The evidence review on which this report is based employed a literature search (quick scoping review) complemented by in-depth interviews with subject experts.

Although claims for Universal Credit (UC) are made on an individual basis, in couple households both partners’ circumstances are taken into consideration in determining the benefit amount. Reflecting the economies of scale associated with living with a partner, the couple rate is lower than the amount paid to single adults. DWP guidance on Living Together as a Married Couple (LTAMC)^{[footnote 2]} is used by decision makers to assess whether unmarried adults living in the same household are a couple. UC payments are usually paid to one adult in the household, although under certain circumstances (e.g. domestic abuse) separate payments may be made to each partner.

An important finding from this evidence review is that there is little evidence on LTFE and how this is shaped by living together policy. Despite the lack of direct evidence, the wider literature points to factors that may affect or contribute to LTFE. Research suggests that living together policy may discourage some people from living with a partner (although there is scarce evidence on whether these factors also contribute to LTFE). Claimants may be concerned about the financial implications of being jointly assessed with a partner, for example whether their income will go down (reflecting the couple rate) and whether they will have full access to benefit income and control over how it is spent (particularly if UC is paid to a partner). However, conclusions about the effects of living together policy on claimant experiences and behaviour should be taken as indicative only, due to the lack of direct evidence on this topic.

Living together policy inevitably reflects assumptions and expectations about who will financially support each other, and in what circumstances. Certain principles underpinning living together policy may be out of step with the behaviour and preferences of some couples. For instance, living together policy is based on the idea that couples pool resources and that financial support in couples is coterminous with living together, but evidence suggests that, increasingly, couples choose to keep their income separate. However, there is a lack of direct evidence on how changing social norms and behaviours contribute to LTFE, so conclusions must remain tentative.

Drawing on the available evidence, this report considers whether there is scope to reform living together policy with a view to reducing LTFE. Reflecting the complexity of the issue and the lack of policy evaluations in this space, the policy discussion is descriptive rather than directive (i.e. no recommendations are made). Reducing reliance on household means-testing represents one option for reform, for instance by taking certain components out of UC and establishing them as separate, individual benefits. However, household means-testing is designed to target support to those most at risk of experiencing financial hardship (who are disproportionately adults living without a partner). Another policy option is to only take a partner’s income into consideration above a certain threshold of individual income. This would in effect mean that the couple rate for UC would only apply to those earning above this threshold. Widening the circumstances in which split payments can be used or making this the default might also help to address LTFE. Finally, it has been suggested that DWP could introduce a transition period in which couples do not have to undergo joint assessment. This would entail an initial period, for example the first six months that the couple is living together, when the partners can still be assessed individually if they choose to be.

Acknowledgements

This research was commissioned by DWP. Grateful thanks are owed to the following experts who took part in an interview (in alphabetical order): Dr Silva Avram, Fran Bennett, Dr Julia Carter, Professor Simon Duncan, Dr Michael Fletcher, Professor John Haskey, Professor Sue Himmelweit, Dr Rita Griffiths, Miranda Phillips, Professor Holly Sutherland. Thanks are also owed to RAND Europe colleagues Lillian Flemons and Lucy Hocking, for their helpful feedback on work produced for this study. All interviewees gave their permission to be named in this report.

The authors

Dr Madeline Nightingale, RAND Europe

Giulia Lanfredi, RAND Europe

Joanna Hofman, RAND Europe

Lucy Gilder, RAND Europe

Dr Chris van Stolk, RAND Europe

Glossary

Benefit error (claimant error): This refers to when a claimant unintentionally claims a benefit they are not entitled to – either because they have provided inaccurate or incomplete information or because they have failed to report a change in their circumstances promptly – but DWP assesses the claimant’s intent as having been not fraudulent.

Source: National Audit Office (2015)^{[footnote 3]}

Benefit error (official error): This refers to when DWP pays the wrong amount because of a lack of action, delay or mistaken assessment by DWP, a Local Authority or Her Majesty’s Revenue and Customs.

Source: National Audit Office (2015)^{[footnote 4]}

Benefit fraud: This refers to when a claimant intentionally claims a benefit they are not entitled to, for example by deliberately providing false information or withholding information.

Fraud occurs when the following three conditions apply:

1. the claimant does not meet basic conditions for receipt of the benefit or the rate of benefit
2. it is reasonable to expect the claimant to be aware of the effect on their entitlement of providing incomplete or wrong information
3. benefit is stopped or reduced as a result of DWP’s review

Source: National Audit Office (2015)^{[footnote 5]}

Benefit unit (or unit of assessment): This refers to the unit that is used for the purposes of calculating benefit entitlement or payment; it could be a single claimant, a couple (joint claimants) or a household (joint claimants).

Source: UK Government (2013)^{[footnote 6]}

Individualisation: A fully individualised means-tested social security system would have the following four main aspects:

• each person would have an individual right to claim financial support, and no one would be able to claim support simply as an adult dependent of another claimant
• assessments of financial need would take place on an individual basis, without taking into account the needs or resources of other adults in the family or household
• the award would cover the needs of that individual only and would not include any payments for adult ‘dependants’
• payments would be made to the individual, so that each individual adult would receive money in their own right

In technical terms, this means that the individual becomes the assessment unit, the resource unit and the payment unit.

Source: Millar (2004)^{[footnote 7]}

Living together fraud and error: This refers to cases where a claimant declares to be single when they actually live with another person and maintain a joint household.

Source: DWP (2020)^{[footnote 8]}

Means-tested benefits (or conditional benefits): This is a type of selective benefit (cash transfers and services limited to individuals or households with limited resources), access to which requires the DWP to check applicants’ resources (i.e. incomes, assets or both).

Source: Gugushvili and Hirsch (2014)^{[footnote 9]}

Universal benefits (or unconditional benefits): These are cash transfers or services that are available to all citizens or residents or to large categories of citizens (e.g. pensioners) without a means-testing requirement or other form of selectivity. The term ‘universal’ encompasses some benefits that to do not go to everybody – they may be demographically targeted or dependent on prior contributions, without being specifically targeted at less well-off households.

Source: Gugushvili and Hirsch (2014)^{[footnote 10]}

Introduction

This report presents findings from an evidence review conducted by RAND Europe on behalf of the Department for Work and Pensions (DWP) about household formation and LTFE in the UK and other OECD^{[footnote 11]} countries.

Living together fraud and error

Benefit fraud is where the conditions for receipt of benefit (or the rate of benefit in payment) are not being met and the claimant can reasonably be expected to be aware of this.^{[footnote 12]} The distinction between welfare fraud and error (Table 1.1) relates to whether behaviour is deliberate or reflects an unintentional mistake on behalf of claimants or officials. Welfare fraud and error can result in overpayment, reducing the money available for other government services, or underpayment, reducing the amount that claimants receive, potentially causing hardship and distress. ^{[footnote 13]} Both fraud and error create administrative costs for departments as they try to identify and correct errors.^{[footnote 14]}

Table 1.1: Definitions of fraud and error in the welfare system

Name	Definition
Fraud	People deliberately declare false information or withhold information in a claim
Claimant error	People mistakenly declare the wrong information in a claim
Official error	Departmental staff make mistakes when checking awards or do not respond quickly in processing information

Source: National Audit Office^{[footnote 15]}

The rate of overpayment and underpayment for all UK benefits and UC is shown in Table 1.2. Across all benefits, overpayment represented 3.9% of expenditure in 2021, compared with 2.1% in 2019. For UC, the overpayment rate was 14.5% in 2021, compared with 8.7% in 2019.

Table 1.2: Rate of fraud and error^{[footnote 16]} (2019 to 2021), in terms of percentage of expenditures overpaid and underpaid

Overpaid	2019	2020	2021
All benefits	2.1%	2.4%	3.9%
Universal Credit	8.7%	9.4%	14.5%

Underpaid	2019	2020	2021
All benefits	1.1%	1.1%	1.2%
Universal Credit	1.3%	1.1%	1.4%

Source: DWP^{[footnote 17]}

LTFE refers to situations where a claimant has a partner but is receiving benefits as a single person. LTFE was the third-largest reason for UC overpayments in 2020 and 2021, as illustrated in Table 1.3.

Table 1.3: Main reasons for Universal Credit fraud overpayments (2021)

	Total overpayments	Overpayments due to fraud
Self-employed earnings fraud	3.8%	3.6%
Capital	2.5%	2.2%
Living together fraud and error	2.0%	2.0%

Source: DWP^{[footnote 18]}

UC overpayments related to living together were higher in 2021 than in 2020 (£760m compared with £215m), as illustrated in Table 1.4.

Table 1.4: Universal Credit overpayments due to living together fraud and error (2020 and 2021)

Overpayments	2020	2021
Amount	£215m	£760m
Percentage	1.2%	2.0%

Source: DWP^{[footnote 19]}

Living together policy

Although claims for UC are made on an individual basis, in couple households, both partners’ circumstances are taken into consideration in determining the benefit amount. Reflecting the economies of scale associated with living with a partner, the couple rate is lower than the amount paid to single adults.

The current DWP definition in legislation classifies couples living together as either:

• two people who are married to, or civil partners of, each other and are members of the same household

or

• two people who are not married to, or civil partners of, each other but are Living Together as a Married Couple

DWP guidance on LTAMC^{[footnote 20]} is used by decision makers to assess whether unmarried adults living in the same household are a couple. UC payments are usually paid to one adult in the household, although under certain circumstances separate payments may be made to each partner. DWP’s living together policy is described in more detail in Chapter 3.

Research questions

The aim of this evidence review is to gather evidence on how the UK’s LTAMC policy affects LTFE and how this relates to such contextual factors as changing social attitudes, gender relations and patterns of household formation. While the review focuses on the UK context, where possible international evidence is used to locate the UK in an international perspective, drawing on examples from other OECD countries.^{[footnote 21]}

This evidence review seeks to answer five research questions:

1. What are the prevailing attitudes and approaches to household formation in the UK and internationally (OECD countries)?

• Do these attitudes impact on the dynamic cycle of household formation and break up?
• Do these approaches impact welfare fraud and error rates?

2. How can a household or benefit unit be defined in social policy?

• What are the criteria for a stable or unstable household unit in the UK?
• What types of households are re-partnering in the UK?
• How does this compare with international policy (OECD countries)? Does this impact on welfare fraud and error rates?

3. What is the relationship between household composition and household finances?

• Is it shaped by socio-economic, geographical or age factors?
• How has this changed over time?
• How does this compare internationally (OECD countries)? Is this reflected in social policy?

4. What is the role of women in the household?

• Do women want, need or seek financial independence from male partners?
• Has this changed over time? And is this represented in UK and international social policy (OECD countries)?
• How does this impact on welfare fraud and error rates?

5. Does the living together policy affect partnering and cohabitation decisions? If so, how?

Methodology

This evidence review is based on a review of literature and in-depth interviews with experts.

Literature review

Two quick scoping reviews (QSRs) were conducted, one focusing on how to define the household or benefit unit for the purposes of social policy and the other on broader social and economic changes: household formation and composition, household finances and money management. QSR was chosen over more systematic approaches to evaluating evidence (i.e. systematic review, rapid evidence assessment) because of the broad scope of the research and the fact that critically appraising the evidence was not a key consideration for this study.^{[footnote 22]} The choice of QSR was also informed by pragmatic considerations about the timeline of the research, which took place between November 2021 and March 2022.

The QSRs focused on sources published in English, relating to the UK or other OECD countries and published between 2011 and 2021. The search strategy is outlined in Annex 1. Additional sources were identified via snowballing,^{[footnote 23]} including some highly relevant sources published prior to 2011. Additional targeted searches were conducted to follow up on areas of interest where fewer sources were identified from the structured search, for instance living together policies in other countries.

After being assessed against inclusion and exclusion criteria, data from included sources were input in a data extraction template in Excel (see Annex 1), enabling systematic analysis against key themes and questions.

In-depth interviews with experts

Interviews were conducted to complement the QSRs and fill any outstanding gaps in understanding of the LTAMC policy, its effects on LTFE or the role of contextual factors. A total of 10 interviews were conducted, in February and March 2022. Interview participants were selected among academic and policy experts in the field of household formation and welfare fraud in the EU and OECD countries. Two main approaches were used to compile a list of potential interviewees:

• literature-based approach: authors of key bibliographical sources deemed particularly insightful during the literature review phase were shortlisted as potential interviewees
• snowballing approach: interviewees were asked for recommendations of other relevant experts to approach to take part in an interview

The list of potential interviewees was approved by DWP before experts were invited to take part in an interview. A total of 10 interviews were conducted to enable data to be gathered from experts with differing backgrounds and expertise, within a small and specialised research community.

Interviews were conducted remotely via Microsoft Teams and lasted approximately 60 minutes each. Interviews were semi-structured to enable comparability while allowing some flexibility to tailor the discussion according to the interests and expertise of the individual. The topic guide for the interviews can be found in Annex 2. Interviews were recorded (with the interviewee’s permission) and the transcripts were analysed thematically.^{[footnote 24]}

Strengths and limitations of this evidence review

This study brings together the available evidence on LTFE in the UK context, providing a novel synthesis of current knowledge. The review was conducted in a structured, comprehensive and transparent way, drawing on evidence published in the past 10 years. The report is informed by interviews with key experts who are leading research and policy analysis in this area.

A number of limitations to this methodology should be noted. While it takes a structured approach, a QSR does not follow the same levels of rigour as a systematic review or rapid evidence assessment.^{[footnote 25]} It is possible that certain relevant sources were missed, particularly those published prior to 2011. The focus on English-language sources may, likewise, have resulted in certain findings being excluded. Although the review focused on the past 10 years, some sources may have been published (or draw on data published) prior to important policy changes, such as the introduction of UC in 2013. In terms of the qualitative research, the study draws on a relatively small number of in-depth interviews; different viewpoints or additional findings might have emerged if the pool of interviewees had been larger.

As an evidence review, the study is constrained by the nature of the evidence available on the topic of household formation and LTFE. Relatively few studies addressing LTFE directly were identified as part of this evidence review, and literature in this area is predominantly comprised of small-scale, qualitative studies. In addition, the evidence gathered approached policy analysis mainly from a theoretical point of view, and there is a lack of empirical studies and policy evaluations in this area. While these studies generate rich data and valuable insights, no impact assessments of living together policy were identified from which to draw causal inferences. Conclusions about the effects of living together policy on claimant experiences and behaviour should therefore be taken as indicative rather than definitive.

Structure of this report

Drawing on evidence from the UK and other OECD countries, Chapter 2 summarises the changing context in which LTFE is situated in terms of household formation and composition, household finances and money management. Chapter 3 focuses on UK policy and how the configuration of the benefits system may affect or contribute to LTFE. Informed by evidence from the UK and overseas, Chapter 4 considers policy options that might be used to address or mitigate LTFE. Chapter 5 draws out key implications of the findings for LTAMC policy.

The changing context of living together fraud and error

This chapter summarises evidence from the UK and other OECD countries on the social and economic context in which LTFE is situated, focusing on patterns of household formation and composition, households at risk of experiencing financial hardship, and the management of money within households, including gendered roles and expectations. The sociodemographic changes described in this chapter set the backdrop for living together policy and LTFE. However, it was not possible to identify any evidence to link these changes directly to the policy context. The data and evidence presented in this chapter are intended to add contextual insights rather than to draw direct lessons about LTFE and living together policy.

Patterns of household formation and composition

Increase in the number of adults living without a partner

Data from the past 10 years show an increase in the number of households in the UK, driven by a surge in the number of people living alone.^{[footnote 26]} These trends are observed across EU member states too. The latest Office for National Statistics bulletin^{[footnote 27]} on families and households^{[footnote 28]} estimated that in 2020 there were 27.8 million households in the UK^{[footnote 29]}, representing a 5.9% increase over the previous 10 years. The average household size in the UK was 2.4 people. ^{[footnote 30]} Overall, the number of people living alone (single-person households) in the UK has been growing at a rate of 4% over the past decade. ^{[footnote 31]} A quantitative study focusing on demographic changes in the UK reported that the proportion of families with dependent children headed by a lone parent has tripled in the past 40 years.^{[footnote 32]} Furthermore, an EU study from 2012 based on an analysis of the European Union Statistics on Income and Living Conditions (EU-SILC) survey data found that the UK and Ireland had the highest rates of lone parents in the EU.^{[footnote 33]}

Data by Eurostat^{[footnote 34]} showed that during the decade 2010 to 2020, the total number of households in the EU increased by 7.2%. In 2020, the average household size in the EU was 2.3 people, and all member states recorded a decrease in the average number of persons per household in the past decade. Single-adult households (i.e. households consisting of only one adult, living with or without children) increased by 19.5% over the decade 2010 to 2020.

Table 2.1: Trends in household formation in the UK and EU (2010 to 2020)

Household formation	UK	EU
Increase in the number of households (2010 to 2020)	5.9%	7.2%
Average household size (2020)	2.4	2.3
Increase in single-adult households (2010 to 2020)	4%	19.5%

Source: ONS^{[footnote 35]} and Eurostat^{[footnote 36]}

Fewer people are choosing to get married, and marriage occurs later in life

A trend that emerged clearly, both in the UK and in the EU, is the increase of cohabitation as a living arrangement. According to the ONS,^{[footnote 37]} the share of unmarried adults living with a partner in England and Wales increased from 11.3% of the population aged 16+ in 2010 to 13.1% in 2020. However, ‘married’ or ‘in a civil partnership’ remained the most common partnership statuses in 2020, accounting for 50.6% of the adult population. In 2020, the highest number of people (in absolute terms) cohabitating without having been previously married or in a civil partnership in the UK was in the 16 to 29 years age bracket (1.98m people), followed by the 30 to 34 years age bracket (997,393 people).^{[footnote 38]} A quantitative study^{[footnote 39]} conducted in the EU based on census data found similar trends to the UK, estimating that in most EU member states, people aged 25 to 34 are most likely to choose cohabitation as a living arrangement.

The ONS estimated that, in 2020, the highest number of people cohabitating while having been previously married or in a civil partnership was in the 55 to 59 years cohort (248,535), followed by the 50 to 54 years cohort (201,569).^{[footnote 40]} According to a quantitative study based on ONS data,^{[footnote 41]} this is part of a broader trend that has seen a progressive move away from the ‘traditional’ family formation (i.e. the nuclear family of a married couple with or without dependent children) and the rise of ‘new’ family forms, such as cohabiting couple families, single-parent families^{[footnote 42]} and blended families.^{[footnote 43]} The same trend was identified also at EU level, mostly driven by changes in societal and family norms and economic contexts – as explained in a recent policy brief.^{[footnote 44]} A qualitative study^{[footnote 45]} exploring the meaning of ‘commitment’ in different types of relationships among a sample of young heterosexual women in the UK found that a committed relationship is not necessarily associated with marriage and that other relationship forms (such as cohabitation or living apart together) can involve equal, if not more, commitment. Nonetheless, research found that cohabitating couples are overall more likely to dissolve in comparison with married couples – as explored later in the chapter. Qualitative social research on social norms around partnership status found that in the UK, cohabitation has begun to take on new meanings, with other indicators of commitment, such as childbearing and shared housing, becoming more important than marriage in defining a relationship.^{[footnote 46]}

Figure 2.1: Marriage rates in England and Wales (1972 to 2018)

Source: ONS (2021)^{[footnote 47]}

The other side of the rise in cohabitation is the decline of marriage − in terms of both the number of people getting married and the cultural role that this institution seems to hold. The British Social Attitude survey showed major changes in marital behaviour in Britain the past 30 years, with growing numbers of people not getting married or delaying this step.^{[footnote 48]} A major trend associated with this phenomenon is couples choosing cohabitation as a precursor to marriage. When the British Social Attitudes survey began, in 1983, the majority of couples (7 out of 10) did not live together before getting married.^{[footnote 49]} Nowadays, marriage without first living together is as unusual as premarital cohabitation was in the 1970s^{[footnote 50]}: ONS data show that in 2018, 88.5% of men and women in opposite-sex couples cohabited before getting married.^{[footnote 51]} A quantitative study on cohabitation and marriage in Britain found that the decline of marriage is especially prevalent among younger cohorts, as recent data from the ONS show.^{[footnote 52]} In fact, Figure 2.1 illustrates how the overall number of married people in England and Wales has decreased over the past four decades. Similarly, a quantitative study^{[footnote 53]} exploring the meaning of cohabitation across Europe through the analysis of the Generations and Gender Programme surveys found that most participants^{[footnote 54]} were reluctant to prescribe a specific time point for marriage. This indicates that the sequence or timing of marriage is no longer normative in most European countries.

Research found that, overall, cohabitating couples in the UK are less stable than married ones. A quantitative study using data from the ONS Longitudinal Study on couples found that 82% of adults aged 16 to 54 who were married in 1991 were living with the same partner in 2001, while only 61% of those cohabitating 1991 remained with the same partner in 2001.^{[footnote 55]} Among those who remained with the same partner, around two thirds had converted their cohabitation to a marriage by 2001.^{[footnote 56]} These figures are consistent with an analysis based on the first wave of the Generations and Gender Programme surveys (2004 to 2011): in the UK, nearly 40% of individuals whose first union was a cohabiting relationship not leading to marriage have dissolved their union.^{[footnote 57]} The percentage of couples dissolving their partnership then drops to slightly more than 10% for individuals who cohabited before marrying, and to less than 10% for those whose first union was a marriage (not preceded by cohabitation).^{[footnote 58]} A study by the Centre for Population Change, based on the Understanding Society Survey, explained that instability of cohabitation is mostly attributed to the sociodemographic circumstances of the couple in this arrangement.^{[footnote 59]} In fact, cohabitants, on average, are younger and have lower incomes − both characteristics that are associated with a higher risk of partnership instability.^{[footnote 60]} For example, data from the The Way We Are Now survey, conducted in the UK in 2016, showed that the biggest external strain on relationships is financial worries, with 26% of respondents experiencing this pressure.^{[footnote 61]}

Figure 2.2: Divorce rates in England and Wales (1972 to 2019)

Source: ONS (2020)^{[footnote 62]}

Note: The ONS dataset provides only disaggregated data for men and women.

The greater dissolution of cohabitating couples is among the reasons that explain why, overall, divorce rates are declining. A quantitative study based on the National Survey of Family Growth in the USA argued that the dissolution of a cohabitating union is, in many ways, an ‘averted divorce’.^{[footnote 63]} This explanation is in line with trends identified in England and Wales by ONS data, where divorce rates have been declining since the early 2000s, as compared with the 1980s and 1990s (Figure 2.2). In older cohorts, people would have faced stronger societal pressures to marry, and these people experienced the same relationship problems that led to marital dissolution.^{[footnote 64]} The pressure to marry has decreased, and people are more likely to cohabit if they are not strongly interested in marriage.^{[footnote 65]} This is visible in ONS data (Figure 2.1), which show that the UK marriage rate has been steadily decreasing since the 1970s.

Consistent with the trends described above, the number of children born out of wedlock or from single parents is on the rise. Data from ONS shows that births outside of marriage in the UK went from 39.5% in 2000, to 46.8% in 2010, to 49% in 2020.^{[footnote 66]} Similarly, in the EU, the proportion increased from 27.3% in 2000 to 42.6% in 2016.^{[footnote 67]} The British Social Attitudes survey observed a concurrent change in social and cultural attitudes.^{[footnote 68]} In 1989, 71% of people agreed with the statement “people who want children ought to get married” and 17% disagreed. In 2013, only 42% agreed, while 34% disagreed with the same statement. A more recent edition (2020) of the British Social Attitudes survey reported that between 1996 and 2017, the proportion of dependent children living in cohabitating households rose from 7% to 15%, and slightly more than one fifth lived in lone-parent families in 2020.^{[footnote 69]}

Research has shown that a growing number of couples in the UK decide to live apart while being together (in a relationship) – especially among the older and higher-income cohorts. This arrangement is in line with the rise of ‘new’ family formations over ‘traditional’ ones. A mixed methods study exploring the living apart together (LAT) phenomenon^{[footnote 70]} explained that a new relationships paradigm is emerging and that family life is no longer equated with the married couple.^{[footnote 71]} Among the reasons mentioned for deciding to live apart while in a relationship, 31% of participants in a mixed methods study conducted in 2013 said that they live apart because the relationship is at an early stage and they do not feel ready to cohabit yet.^{[footnote 72]} At the same time, 30% of participants expressed a preference for not living together due to wanting to keep their own home, prioritising other responsibilities over the relationship (including own children), and ‘just not wanting to live together’.^{[footnote 73]} Furthermore, 12% reported geographical constraints, such as their partner having a job or studying elsewhere or living in an institution (i.e. a nursing home or prison).^{[footnote 74]} The active decision to live apart while being together seems to be perceived as holding considerable benefits, especially among older cohorts.^{[footnote 75]}

A review of the literature^{[footnote 76]} on couples found that among the main advantages of this arrangement is an increase in autonomy for women (who in this way can avoid being burdened with additional unpaid domestic work); the enhanced ability to manage relationships with own children, parents and friends; and the reduced risk of asset depletion in the face of a relationship breakdown. This last element was identified by another study (based on the analysis of the Household, Income and Labour Dynamics survey) as the main reason why 73% of the older (over 45), previously married cohort in Australia had made a definite decision to live apart.^{[footnote 77]} Further quantitative research conducted in the UK on the phenomenon found that older individuals and those who have experienced divorce or widowhood are more likely to be in a LAT union than married, mostly for reasons of independence.^{[footnote 78]} Conversely, younger people are more likely to say that they are not ready to live together.^{[footnote 79]} Overall, though, living apart seems to be less feasible for low-income couples, because having fewer financial resources makes it a necessity to share household expenses (i.e. housing, food, bills), and the term LAT itself seems to be associated more with the middle class.

Decisions about household formation and composition are partly influenced by economic and financial factors

This section presents evidence on how economic and financial factors have been found to play a role in shaping decisions on household formation. This is especially relevant in the context of the LTAMC rule discussed in Chapter 3, which determines that two individuals living together (in a sexual and romantic partnership) are to be assessed jointly for benefit purposes.

In some European countries, the high social and economic cost of divorce acts as a deterrent to the dissolution of households. One quantitative study found that in Italy and Poland, both characterised by strong (Catholic) religious views, the process of getting a divorce is often lengthy and expensive, and this discourages couples from taking this path.^{[footnote 80]} Another element that acts as a deterrent is women’s economic dependence on men, which results in some women struggling to separate from their spouse due to a lack of means to support themselves financially.^{[footnote 81]} For these reasons, separation is more than just a short-term transition to divorce, as many separated couples avoid official divorce proceedings and remain legally married despite being de facto separated.^{[footnote 82]} All these factors figure as drivers of low divorce rates.

Living in a welfare state influences marriage and fertility rates, due to the ways in which public policies (i.e. benefits, tax deductions) provide financial support to families. A quantitative study of OECD countries found that the amount of public social spending in support of families has an impact on trends of family formation.^{[footnote 83]} For example, living in a generous welfare state (i.e. a welfare state with larger public social spending) increased both marriage and divorce rates, with a stronger effect on the former.^{[footnote 84]} The research also found a positive association with fertility rates, especially among non-married couples. The authors’ argument is that this association happens because welfare states tend to subsidise births by providing benefits and tax deductions in support of families with children. The authors also explained that welfare states take on some of the financial burden of the support functions that would traditionally be provided within the family − such as care for children and older people, and support in case of unemployment and illness. Consequently, they concluded, welfare states support family formation.^{[footnote 85]}

Conversely, when insufficient support is provided by the state, intergenerational households can afford protection from financial hardship. Quantitative research conducted in the EU and the USA shows that, while in the past households composed of multiple generations were meant to provide care for their elders, now families that choose this living arrangement tend to do so to draw on the financial support their elders can provide.^{[footnote 86]} In fact, this type of co-residence is generally opted for by groups that experience socio-economic disadvantage. It is also more prevalent among women, the widowed, those with lower education, those without paid employment and those from a migrant background. In Romania, for example, the decision to live with grandparents as a coping strategy for financial hardship is driven by two factors: (i) the high number of parents who have migrated abroad for work and have left the extended family to look after their children; and (ii) the system of social assistance benefits being largely tied to earnings, which therefore favours middle-class working parents.^{[footnote 87]} Thus, lower-income and jobless families, as well as those with irregular work histories, tend to rely more on relatives for financial support.^{[footnote 88]}

Financial considerations also influence decisions about cohabitation. A mixed methods study exploring the phenomenon of intermittent cohabitation in the USA reported that women in this arrangement did so primarily for practical motives.^{[footnote 89]} In particular, the financial dimension was found to be particularly important for mothers, who often decided to cease cohabitation if a partner no longer contributed adequately to the household, regardless of romantic feelings.^{[footnote 90]} Similarly, housing policy was found to strongly determine mothers’ ability to cohabit.^{[footnote 91]} The article explained that public housing vouchers in the USA generally include only the mother and her children, and there are specific provisions regulating who else is allowed to stay in the home. Therefore, mothers reported undergoing involuntary separations when they felt under scrutiny, for fear of losing the entitlement.^{[footnote 92]} Housing benefits in the UK are now part of UC, but one interviewee remembered encountering a similar situation when conducting research on housing benefits under the ‘legacy’ system.^{[footnote 93]} They mentioned the episode of a woman whose husband was posted overseas and was not financially supporting her. She was therefore on housing benefits. When he came home, he could not stay at his parents’ house, though, and ended up staying with his wife, who was later prosecuted for benefit fraud.

Households at risk of experiencing financial hardship

Figure 2.3: Rate of people at risk of poverty in the UK, by household type (2011 and 2018)

	2011	2018
Single person with dependent children	33.5%	39.8%
Two adults with three or more dependent children	21.5%	32.5%
Single female	28.2%	28.6%
One adult 65 years or over	29.2%	28.0%
Single person	25.4%	26.7%
One adult younger than 65 years	22.1%	25.5%
Single male	21.8%	24.5%
Households with dependent children	16.5%	21.0%
Two or more adults with dependent children	14.1%	18.1%
Two adults, at least one aged 65 years or over	17.7%	16.3%
Two adults with two dependent children	12.1%	16.2%
Households without dependent children	15.9%	16.1%
Three or more adults with dependent children	11.9%	14.5%
Two adults with one dependent child	13.3%	13.7%
Three or more adults	9.0%	12.6%
Two adults	13.8%	12.4%
Two or more adults without dependent children	12.4%	12.4%
Two adults younger than 65 years	10.8%	9.1%

Source: Eurostat (2022)^{[footnote 94]}

Adults living without a partner (alone or as a single parent) face an elevated risk of poverty

Quantitative evidence shows that poverty is concentrated in smaller households, such as single-person households (especially single pensioners) and single-parent households (especially single mothers) (Figure 2.3).^{[footnote 95] [footnote 96] [footnote 97] [footnote 98]} Couples, with or without children, on the other hand, are less exposed to the risk of poverty.^{[footnote 99]} According to a recent study, this is partly due to the way in which welfare provision is directed at families with children, leaving single-person households with less support from the state.^{[footnote 100]} Furthermore, experts interviewed as part of this study pointed to the fact that living together provides some financial advantages, such as sharing housing costs and economies of scale. Quantitative research found that in all European Union countries,^{[footnote 101]} the relative risk of a child being poor is higher in lone-parent households (47% of families in this group are at risk of poverty) than in two-parent households (21% of families in this group are at risk of poverty).^{[footnote 102]} However, solo living is more strongly associated with poverty in Nordic and north-western European countries (including the UK) compared with eastern and southern European countries.^{[footnote 103]} According to Eurostat data from 2017, the UK had the third highest number of single adults with dependent children at risk of poverty in the EU (almost 60%), behind Bulgaria and Ireland.^{[footnote 104]} Additionally, the risk of experiencing income poverty is more strongly positively associated with having two or more children in the household (rather than one), as well as with the lone parent being single or separated (rather than widowed or divorced), the single parent being a woman (rather than a man), not having upper secondary education, and not being in full-time employment.^{[footnote 105] [footnote 106]}

Female poverty stands out as a relevant pattern across Europe. Results from a multilevel regression study showed that children are significantly less likely to be poor if they live with lone fathers rather than with lone mothers, everything else being equal.^{[footnote 107]} A quantitative study conducted in the UK confirmed this pattern, and illustrated that 60% of lone mothers experience poverty, versus 30% of lone fathers.^{[footnote 108]} Furthermore, as shown in Figure 2.3, single female adults are, overall, poorer than single male adults in the UK.

The configuration of the benefits system can also contribute to financial hardship

Households relying on benefits in the UK may be exposed to financial hardship due to the practical implications of policy design. A mixed methods study exploring the provision of UC in the UK found that this benefit was not designed to reflect changes in family circumstances in a dynamic way.^{[footnote 109]} Entitlement to UC depends on the income and earnings of both members of a couple (for as long as they cohabit) and on family needs and costs (i.e. housing and childcare). This makes it more difficult for claimants to budget for the upcoming month if they have experienced a change in family circumstances. Low-income families, though, often face multiple and unexpected alterations in type or amount of income, housing costs and household composition – which all have an impact on the amount of payment they are entitled to receive. Some of these changes are inherently short-term and unstable, such as children in separated families living at intermittence with either parent, and the system is slower to adapt to variations in terms of family composition, in comparison with variations of other circumstances (i.e. wages) – as the report explained:

With UC monthly assessment, if claimants report a change of circumstances before their assessment date, it will apply for the whole of the month prior to that date. Hence UC is likely to be more responsive to wages (via the Real Time Information system) than other changes, and so may be unpredictable, making it more difficult for claimants to budget. UC was intended as a ‘dynamic’ benefit in relation to the labour market, focused on changing ‘pro-work social norms’, rather than dynamic in relation to family circumstances.”^{[footnote 110]}

In fact, as described in Chapter 3, UC inherited the household as the unit of assessment from the previous, ‘legacy’ system.

Gender equality and household income

Increasingly, couples choose not to pool all financial resources

Social research has found that the ‘unitary model’ of the household (see Box 2.1) is no longer well aligned with people’s preferences and behaviour in terms of managing household finances. According to an EU-level study based on EU-SILC data, a significant proportion of adults reported keeping at least some of their personal income separate from that of their partner, and overall fewer than half (45%) of households were fully pooling resources.^{[footnote 111]} Similarly, data from the 2012 Poverty and Social Exclusion Survey in the UK showed that 49% of couples pooled and managed finances jointly (with a further 15% pooling some of the money and keeping the rest separate), and 23% stated that one member of the couple is responsible for all the household money except for the other partner’s personal spending money.^{[footnote 112]}

Quantitative evidence also showed that, although the majority of couples do not pool all their income, the unitary model (see Box 2.1) is still more likely to be true for couple households (compared with other types of multi-person households).^{[footnote 113]} Furthermore, a traditional division of work between spouses (where the man is the sole breadwinner and the woman is a full-time homemaker) increases the likelihood of full income pooling, while dual‑earner couples and unmarried couples are less likely to report full income pooling^{[footnote 114] [footnote 115]} (even more so when the woman’s earnings represent more than 30% of the couple’s earnings^{[footnote 116]}). Other factors positively associated with full income pooling for couples are having dependent children together, being retired (suggesting a generation effect but also, perhaps, signalling lower incomes), being in a long-term relationship and being from older demographic cohorts.^{[footnote 117] [footnote 118]} In contrast, people who are financially tied to other households (e.g. blended families^{[footnote 119]}) are less likely to pool resources, as are higher-educated and higher-income couples, and couples in the early stages of their relationship.^{[footnote 120] [footnote 121]}

On the whole, research estimated that, considering the trends of decreasing marriage and increasing cohabitation, as well as the increasing proportions of dual-earner couples,^{[footnote 122]} the share of ‘full pooling’ households is likely to continue decreasing.^{[footnote 123]} as one interviewee pointed out, socio-cultural norms around finances are becoming increasingly more individualised, and overall, at societal level, there is a greater desire for individual resources, even when people are part of a couple.

Box 2.1: Economic models explaining the division of household finances

Economic theory has identified two major ways in which people who are part of the same household use their financial resources to support their living.

• The unitary model^{[footnote 124]}: In this approach, the household is assumed to behave as a single entity, a rational con­sumer maximising a unique utility function under a single budget constraint. For households to func­tion ‘as if’ they were individuals, two main assumptions have to be met: (i) individual preferences have to converge, so that the house­hold can be considered a single decision unit; and (ii) household members’ resources have to be pooled and (enhanced by economies of scale) subjected to only one budget constraint. The fact that individuals’ preferences may diverge is overlooked, and issues of intra‑household distribution (i.e. inequalities among members of the household) are bypassed. Individuals are not discernible within the household, which operates as a ‘black box’. No account is taken of the possibility that some members of the family are rich while others are poor. Individuals’ eco­nomic well‑being is conventionally measured on the basis of household-level information and is assumed to be the same for all household members.
• Alternative, non-unitary models^{[footnote 125]}: Non‑uni­tary models consider that each household member (most models consider two decision makers) has their own utility function, and incomes are not as­sumed to be pooled. Estimating these models, though, is also quite complex and not easy to operate for statistical purposes. There seems to be consensus that applying a unitary model logic overlooks inequalities present within the household (in terms of distribution of resources) but that operationalising alternatives is not a simple task. For example, understanding how and to what extent resources are pooled requires an in-depth understanding of the individual household situation, which is not always attainable with statistical computations.

Women are more concerned than men about autonomy and independence around household finances. Both the literature and interviews highlighted that pooling income into a joint bank account does not necessarily entail joint management of finances or equal power.^{[footnote 126]} For example, a quantitative study examining intra-household distribution of consumption in 12 countries found that equal sharing rarely happens, creating gender inequality in consumption and poverty, with men benefiting from a larger share of resources.^{[footnote 127]} Abusive partners may also force their partners to use a joint bank account, and this financial dependence can prevent the victim from seeking support or make it more difficult for them to do so.^{[footnote 128]} Overall, qualitative research found that women seem to value the access to an independent income more, either from wages or benefits, as compared with men.^{[footnote 129]} This reveals a desire for individual control over financial decision making within the household, and for being able to spend money in autonomy without having to justify this to a partner.^{[footnote 130]} Qualitative research found that women in low-income households in the UK expressed a preference to maintain a certain degree of autonomy from the partner.^{[footnote 131]}

UK policy and living together fraud & error

This chapter introduces DWP’s living together policy and summarises the evidence on how this policy may affect LTFE.

Defining the household or benefit unit

The UK relies to a great extent on household means-testing

Welfare systems in different countries rely on targeted (rather than universal) benefits to varying degrees,^{[footnote 132]} and they take different approaches to this. The unit of assessment (or benefit unit)^{[footnote 133]} can be an individual or a household.^{[footnote 134]} In most OECD countries, the unit of assessment is the household or the family, usually defined in terms of the nuclear family, i.e. the spouse or partner and any dependent children.^{[footnote 135]} However, there are different ways of defining the household or family for the purposes of social policy. In some OECD countries (Austria, Japan and Luxembourg), other non-dependent adults who live together (i.e. those who do not have a romantic or sexual relationship) form part of the benefit unit.^{[footnote 136]} In some OECD countries (Austria, Germany, Japan and Switzerland), financial support from family members outside the household – parents, grandparents and even siblings – is taken into account when determining the level of benefit.^{[footnote 137]} This cross-national variation reflects assumptions and expectations about who will financially support each other, and in what circumstances. Within the same country, the unit of assessment may differ across different types of benefits. This is the case in the UK, which, as described by one interviewee, does not have – and has never had – a single benefit unit; it varies across different benefits. However, the introduction of UC in the UK means that a larger proportion of social welfare is directed towards a single benefit unit.

This evidence review identified relatively few sources addressing LTFE in countries other than the UK. LTFE may be less prevalent, or the issue may be less high on the policy plan in countries where there is greater reliance on contributory benefits and means-tested welfare occupies a more residual position (there is less scope for LTFE where there is greater reliance on the household as the unit of assessment). Administrative factors, such as the existence of comprehensive population registers (which do not exist in the UK), may also make LTFE less prevalent in other countries, although this could not be established from the literature. The emphasis placed on LTFE in the UK may also reflect the relatively large number of single-parent households in the population.^{[footnote 138] [footnote 139]} International evidence is predominantly drawn from Australia and New Zealand (see Box 4.2 and Box 4.3: in Chapter 4), two countries which share with the UK a strong emphasis on means-tested welfare.

In the UK, a key element of targeting is the household means-test, which has come to occupy a more central position in the benefits system over time.^{[footnote 140]} The rationale for targeting based on household income is to direct resources at those most at risk of experiencing financial hardship. Targeting based on household income also accounts for economies of scale associated with living with a partner.

Unmarried couples form a single benefit unit if they are Living Together as a Married Couple

Under the current system, UC claims are made by individual claimants, and, where relevant, this claim is linked to a claim from a partner. In couple households, both partners’ circumstances will be taken into consideration when determining the benefit amount.^{[footnote 141]} For the purposes of assessing eligibility for UC, DWP guidance^{[footnote 142]} states that two people should be considered a couple if they live in the same household and are one of the following:

• married to each other
• civil partners of each other
• living together as if they were married

DWP guidance states that the household should be given “its normal everyday meaning … It [the household] is a domestic establishment containing the essentials of home life”.^{[footnote 143]} According to this guidance, members of a household share a dwelling,^{[footnote 144]} and both members should live there regularly. In addition to sharing a dwelling, other factors that may be considered in assessing whether individuals form a joint household include:^{[footnote 145]}

• the circumstances in which the two people came to be living in the same house
• the arrangements for paying for the accommodation
• the arrangements for the storage and cooking of food
• the eating arrangements (whether separate or not)
• the domestic arrangements, such as cleaning, gardening and minor household maintenance
• the financial arrangements (who pays which bills? is there a joint bank account? whose name is shown on utility bills?)
• evidence of family life

If they are not married or in a registered partnership, two individuals who are members of the same household need to be LTAMC to be considered a couple and therefore subject to joint assessment.^{[footnote 146]} The LTAMC rule (known until 2005 as Living Together as Husband and Wife) exists to ensure that people who are living together and not married are not treated any more favourably than people who are married (see Box 3.1). The term Living Together as a Married Couple is not defined in legislation.^{[footnote 147]} Decision makers must determine on the basis of available evidence whether the relationship of two people who are not married to each other is comparable to that of a couple who are married.

The LTAMC definition is complex because it considers several different dimensions, not all of which have to be present. According to the guidance,^{[footnote 148]} “Marriage is where two people join together with the intention of sharing the rest of their lives. There is no single template of what the relationship of a married couple is. It is a stable partnership, not just based on economic dependency but also on an emotional relationship of lifetime commitment rather than one of convenience, friendship, companionship or the living together of lovers”. Various characteristics of the relationship may be considered, including mutual love, faithfulness, endurance, interdependence and devotion.^{[footnote 149]} Other factors taken into consideration are the management of household finances (“in most marriages it would be reasonable to expect financial support of one partner by the other, or the sharing of household costs”), the stability of the relationship (including whether such activities as shopping, cooking, cleaning and caring are undertaken jointly), having children, public acknowledgement of the relationship (for instance, in claiming benefits, from friends or family members or sharing a surname) and shared plans for the future.^{[footnote 150]} A signed statement or letter from the claimant that they are LTAMC or intend to marry or enter into a civil partnership is considered sufficient evidence of LTAMC.^{[footnote 151]} Due to the complexity of the LTAMC policy, claimant guidance refers simply to “a partner moving in”.

Box 3.1: Historical evolution of the cohabitation rule

The LTAMC policy has been in effect since 2005. Prior to this date, a similar policy existed, known as Living Together as Husband and Wife (LTAHAW), dating back to 1977. Following the introduction of civil partnerships, in April 2005, the LTAHAW rule was further extended to same sex couples and renamed to reflect gender-neutral language.

Source: Griffiths (2016)^{[footnote 152]}

Continuity between Universal Credit and the ‘legacy’ system

The emphasis on household means-testing in the current UK benefits system is not new. UC – introduced in 2013 – replaced six means-tested benefits.^{[footnote 153]} Although it represents a radical change in creating a single, unified system, UC maintains the focus on the household as the unit of assessment. In the words of one interviewee, “fundamentally, the way in which a household unit is conceived [in UK social policy] hasn’t changed”. The rule is the same under UC as it was under the previous, ‘legacy’ system. However, one notable change under UC is that benefit is paid to one adult on behalf of the household by default. Split payments are possible in certain circumstances (domestic abuse, serious financial mismanagement), but these alternative payment arrangements (APAs) are used infrequently, and generally are only on a temporary basis.^{[footnote 154]}

Implications of the benefit unit for living together fraud and error

There is relatively little research on the topic of living together fraud and error

Only two of the sources identified in this evidence^{[footnote 155] [footnote 156]} review describe living together fraud or welfare fraud.^{[footnote 157]} Other sources acknowledge that the phenomena they describe could contribute to what would for official purposes be deemed fraud but state that evaluating or assessing this falls outside their remit.^{[footnote 158]} Other sources reject the term fraud on the grounds that such behaviour is often inadvertent or accidental (i.e. error)^{[footnote 159]} or that the discussion around welfare fraud often fails to recognise the complex reasons why it occurs.^{[footnote 160]} One interviewee acknowledged that this can be a grey area: people may deliberately downplay the existence or seriousness of their relationship, but this does not constitute fraud if they genuinely maintain separate households:

I wouldn’t call it fraud, because if somebody is living in a separate household, then that isn’t fraud. It [fraud] is only when partners are living there, quite clearly, and don’t have alternative households and so on. So, in terms of those unintended consequences of women living apart from the partner, I would call it ‘living apart together’, which is what it’s called in the sort of middle class, well not just in the literature. But people do it right throughout the income spectrum, don’t they, but it’s called different things depending on whether you’re claiming benefits or not, so I wouldn’t call it fraud.

The distinction between fraud and error also constitutes a grey area. In practice, it may be difficult to distinguish between deliberate actions and inadvertent mistakes and misunderstandings (which is why this report refers to ‘living together fraud and error’ rather than ‘living together fraud’).^{[footnote 161]} Despite the lack of direct evidence on welfare fraud, the literature points to mechanisms that may plausibly affect or contribute to LTFE. It is important to recognise, however, that these relationships are often not directly tested or evaluated. Conclusions about the effects of living together policy on claimant experiences and behaviour should be taken as indicative rather than definitive.

The couple rate may create financial disincentives to living with a partner

The couple rate (i.e. the lower benefit rate paid to claimants living with a partner compared with those who are single), means that some households may be worse off financially if they live together compared with if they live apart. According to analysis by the Institute for Fiscal Studies^{[footnote 162]} the majority of couple households in the UK (68%) would be better off if they were living apart,^{[footnote 163]} rising to 95% of couples with children. However, another study (which takes into consideration economies of scale associated with living with a partner) suggests greater variation in the financial implications of living together as a joint household.^{[footnote 164]} This source finds that financial disincentives to living with a partner are greater for working families (particularly dual-earner families) compared with families where both partners are unemployed or economically inactive.^{[footnote 165]} However, there is a lack of up-to-date evidence on the financial implications of living with a partner (and how this is shaped by the benefit system). The two studies cited above were published in 2010 and 2012, prior to the introduction of UC. Compared with the previous system, UC is designed to direct more resources to couples (particularly those with children) relative to single adults.^{[footnote 166]} A large proportion of couple households with children have seen an increase in their entitlement under UC compared with the previous system.^{[footnote 167]} In light of this, further research is needed to establish whether there are financial disincentives to living with a partner under UC.

Regardless of the economic realities, research highlights a perception among benefit claimants that they may be worse off living with a partner compared with living alone.^{[footnote 168] [footnote 169] [footnote 170]} Griffiths found that the expectation of loss or reduction in entitlement to benefits associated with moving in with a partner could deter low-income mothers from taking this step.^{[footnote 171]} Some women who took part in this research study commented that they would be worse off financially under joint assessment with a partner compared with living as a single mother. In the words of one of claimant interviewed as part of this research study:

I work part time and … get Working Tax Credits … Housing Benefit and help towards childcare costs.… That gets me by.… I’d lose all that if I moved in with him.… We’d have to live on his wage if we were living together … we’d have hardly anything to live on.… We struggle as it is … so we just can’t do it at the moment”. (Miriam, 23, one child aged 18 months)^{[footnote 172]}

Financial support may extend beyond couple relationships, and couples may share resources and support one another financially even if they do not live together. Claimants interviewed as part of one study commented on how economies of scale may be present in other types of households that are not considered to form a benefit unit and subject to joint assessment.^{[footnote 173]} In the words of one claimant interviewed by Griffiths et al.:

[When we first moved in together] I wasn’t entitled to Universal Credit at all, but I was when I was living with my parents, so that made no sense to me.… I don’t get, like, why people who live with their parents are entitled to it, more than … someone who lives with their partner … because when you live with your partner … he shouldn‘t be, like, fully, like, responsible for me because we’re living together”. (Isla, joint claimant, female, Cumbria, single-earner couple, two children)^{[footnote 174]}

Interviewees discussed the possibility of couples who are LAT (who are not considered to form a household for the purposes of joint assessment) financially supporting one another. LAT couples may be considered couples according to other aspects of a relationship considered by decision makers, for instance public recognition of the relationship:

Most people living apart together in our sample wouldn’t be seen as living together in those terms. Some of those terms would apply, like being recognised by others, or acting as a sort of couple in a social sense. But of course most of them had different houses and different bank accounts and that sort of thing.

Financial support may also be offered by other (resident or non-resident) adults, such other family members, particularly in more complex family forms:

Extended families living within a single domestic unit might provide … financial support for a nuclear family within that that’s not recognised by the system at all.

Research highlights low awareness and understanding of the rules around cohabitation and benefit entitlement. Qualitative research from the UK suggests that many people living on low incomes are unaware of how their household circumstances affect eligibility for means-tested benefits.^{[footnote 175] [footnote 176] [footnote 177]} For instance, several of Griffiths’ research participants were surprised to find themselves better off after splitting up with a partner.^{[footnote 178]} There is low awareness and understanding of the LTAMC rule and how it is applied; for instance, there is a widespread myth among claimants that staying over for at least three nights a week constitutes LTAMC. This suggests that the financial implications of moving in with a partner – or disclosing such a move to the authorities – may not be a central factor in decision making, even for those for whom there may be negative financial implications.

It has been stressed in the literature^{[footnote 179]} that claimants’ decision making is not purely rational and motivated by self-interest. As one interviewee argued, individuals do not act with a strict economic rationality, motivated simply to take the course of action where they will be financially better off:

My worry is that [DWP] are focusing … [on] decision making as being a rational economic choice when it’s actually more of a moral and social choice”.

Research on couples LAT indicates that entitlement for benefits can be a motivating factor for a small proportion (around 1%) of such couples, although the evidence base is not robust enough to draw firm conclusions about this. Although there is some evidence to suggest that older couples sometimes decide to LAT to preserve pension entitlements or to avoid taking on a partner’s debt, experts commented that financial considerations, particularly relating to means-testing and benefits, were rarely mentioned by their research participants in relation to decisions about partnership and household formation. There was also some scepticism among interviewees about the idea that benefit entitlement was a key factor affecting partnership decisions and household formation:

People will not make major decisions in their lives because of a benefit. It might temporarily play some role, but they are not going to decide ‘I’m going to marry someone’ just to get a benefit or ‘I’m going to divorce someone’ just to get a benefit.

Similarly, while financial stability and independence are important factors, most people on low incomes reject the idea that decisions about partnership are influenced by benefit entitlement.^{[footnote 180] [footnote 181]} Research suggests that many claimants do not want to be dependent on benefits. Recent research with couples claiming UC found that claimants “wanted to escape the constant scrutiny, their feeling of a lack of control, the fluctuations in income, and the time and effort involved in managing their [UC] claim”.^{[footnote 182]} Furthermore, people on low incomes are acutely aware of the severe consequences that could befall them if they are caught committing welfare fraud. ^{[footnote 183] [footnote 184] [footnote 185]} Some interviewees talked about the challenges of conducting research in this area because stigma associated with welfare fraud may make people hesitant to discuss anything that could be considered potentially fraudulent.

In sum, the couple rate, or at least an expectation of being worse off when living with a partner, may affect decisions about partnership and household formation (although there is no evidence to link this directly to LTFE). However, financial considerations do not exist in a vacuum; they sit alongside a range of other factors in shaping decision making in this area.

Partnership decisions may be driven by concerns about financial independence and autonomy

Research highlights the high value placed on financial independence and autonomy in relationships. Couples who LAT are a diverse group, with varying circumstances, but a desire for financial independence is identified as a key factor motivating decisions to LAT. Qualitative studies highlight the importance of financial independence and autonomy to people on low incomes specifically,^{[footnote 186] [footnote 187] [footnote 188] [footnote 189] [footnote 190]} particularly those who have experience of controlling or abusive relationships.^{[footnote 191]} As described in Chapter 2, financial independence is particularly important for women,^{[footnote 192]} who are more likely to be responsible for day-to-day management of household finances and who are more likely than men to have experienced financial abuse.^{[footnote 193]}

Qualitative research shows how concerns about financial independence and autonomy can influence decisions about partnership and household formation. ^{[footnote 194]}, ^{[footnote 195]} Griffiths’ research, conducted with 51 low-income mothers in the UK, found that in some cases women chose not to move in with a partner because they were concerned about the loss of financial independence associated with living as a couple, including undergoing joint assessment for benefits.^{[footnote 196]} Conversely, not having (full) access to joint income could be a factor contributing to the decision to end a relationship.^{[footnote 197]} For Griffiths’ research participants:

Rather than simply the absolute value of household income or the aggregate monetary value of benefits to which a household may be entitled, more influential in decisions affecting partnering dissolution was thus the extent to which the female partner had access to, and some level of control over, the family’s income and benefits.”^{[footnote 198]}

Financial independence and autonomy are particularly important for people who have reasons to doubt the reliability of their partner, for instance if their partner is over-indebted; has a history of insecure employment; or has issues with drug, alcohol or gambling addiction.^{[footnote 199] [footnote 200] [footnote 201] [footnote 202]} As a broad trend, people tend to partner with individuals who have a similar educational and income profile to them – a phenomenon known as ‘homogamy’.^{[footnote 203]} This means that many benefit claimants, who tend to be on the lower end of the income distribution, may be considering moving in with a partner who does not offer strong prospects for financial stability. For Griffiths’ research participants, being assessed as a single parent offered greater financial stability than being jointly assessed with an unreliable partner:

Although lone motherhood was not without its own challenges and risks, the financial safety net provided by the welfare state was perceived by these mothers to offer a better chance of security and family stability than becoming dependent on an unreliable ‘breadwinner’ or on a new, unproven partner.”^{[footnote 204]}

After conducting interviews with benefit claimants who have failed to disclose a partner to the authorities, Kelly observed that in several cases, this failure was because the partner was perceived to be unreliable or unpredictable.^{[footnote 205]} People may also be reluctant to rely financially on or expose their children to a new partner whose reliability has not yet been established.^{[footnote 206]} In instances when the new partner is not the (biological) parent, there may be concerns about whether they will expect to financially support children in the household, or whether they will do so in practice, particularly if they are paying child support to children in a another household from a previous relationship.^{[footnote 207]} Uncertainty about relationships is a key factor affecting decisions about household formation^{[footnote 208] [footnote 209] [footnote 210]}; when people are still testing out the stability of a relationship, joint assessment raises the stakes associated with committing to and moving in with a partner. In the words of one expert interviewed as part of this study:

There’s such a huge risk on the part of [claimants], particularly when lone parents were re-partnering with somebody who wasn’t the father of their child. To actually give up all of your financial independence and all your own benefits and expect to be supported by a new partner that you haven’t potentially lived with before is a massive ask”.

Protecting benefit income for dependent children may be a particular concern under UC since benefits are not earmarked.^{[footnote 211]} Historically, there has been a drive to label specific benefits as targeted for children (e.g. Child Benefit,^{[footnote 212]} Child Tax Credit^{[footnote 213]}) and directing these benefits towards mothers, with a view to increasing the likelihood of that money being spent on children.^{[footnote 214]} Although child benefit remains a separate benefit, Child Tax Credit has been replaced by UC, and low-income mothers may be concerned that the portion intended for children may be ‘lost’ if their partner has full or partial control over this money.

Under UC, claimants are jointly liable for benefit repayments, including those incurred by their partner under a previous claim.^{[footnote 215] [footnote 216] [footnote 217]} In signing up for a joint claim, claimants take on any debt owed to the government by their partner. This may deter claimants from moving in with a partner or disclosing a partner to the authorities. This was not addressed directly in any of the identified sources, although one interviewee mentioned that a desire to avoid taking on a partner’s debt came up in their research as a factor motivating the decision to live separately from a partner.

Particularly when paid to only one member of the household (as is the default under UC), benefit income may not be shared equally among household members. One partner may have greater control over household resources and restrict the other partner’s access to benefit income, even in cases where UC is paid into a joint bank account.^{[footnote 218] [footnote 219] [footnote 220]} This could lead to financial or material hardship for the other partner, even in households not officially counted as living under the poverty line.^{[footnote 221]} Research suggests there is a gender dimension to this issue, where men are more likely to exert disproportionate control over household resources, including income from benefits.^{[footnote 222] [footnote 223]} One source described this as “enforced financial dependency”,^{[footnote 224]} since claimants may find themselves reliant on their partner to share benefit income equally or fairly. Qualitative research highlights how financial dependence can adversely affect relationship dynamics, for instance making the partner who contributes less financially feeling less able to speak up and assert themselves.^{[footnote 225] [footnote 226] [footnote 227] [footnote 228] [footnote 229]} Having access to independent income may give the individual more ‘say’ in household finances.^{[footnote 230]} In the words of one source based on interviewees with UC claimants:

Having to ask the other partner for money could change the relationship dynamic and undermine a sense of equality; it was described by some as demeaning and infantilising, though partners might not realise this”.^{[footnote 231]}

Concerns have been raised that assessment may increase the risk of financial and other forms of domestic abuse

Some of the identified sources expressed concerns about the potential for joint assessment – particularly when combined with single payment – to place claimants and their children at greater risk of domestic abuse.^{[footnote 232] [footnote 233] [footnote 234]} Domestic abuse is a widespread and gendered issue in the UK, affecting an estimating 1.6 million women per year.^{[footnote 235]} Financial abuse may affect up to one in five adults, disproportionately affecting women on low incomes.^{[footnote 236]} Financial abuse involves “a pattern of controlling behaviour in which one partner seeks to control the other’s ability to acquire, use and maintain financial resources”.^{[footnote 237]} Given historic gendered expectations about financial control and money management described in Chapter 2, financially abusive behaviour could in some instances be inadvertent, reflecting culturally conditioned preferences and expectations, whereas in other instances it may be driven more directly by a desire to control and disempower a partner.

Concerns have been raised in the literature that joint assessment (and payment) may enable financial abuse,^{[footnote 238]} but further research is needed to substantiate this. Qualitative research with UC claimants highlights examples of benefit income being redirected from much-needed household costs and used by a partner to pay off debts or fund an alcohol or drug dependency.^{[footnote 239]} The single payment system consolidates control over finances and makes it harder for the other partner to strengthen their financial capabilities.^{[footnote 240]} An abusive partner may restrict access to jointly assessed income, placing the other partner at risk of experiencing material or financial hardship and making it more difficult for them to leave the relationship. ^{[footnote 241] [footnote 242] [footnote 243]} In this way, joint assessment and single payment may contribute to other forms of domestic abuse (e.g. physical, psychological and sexual abuse), which are frequently present alongside financial abuse.^{[footnote 244]} However, it is important to note that while a link between abuse, on the one hand, and joint assessment and single payment, on the other, has been theorised in the literature, to date there is no empirical research to substantiate this.

The risk of financial abuse associated with joint assessment and single payment is recognised by claimants, as highlighted by one study based on interviews with UC claimants.^{[footnote 245]} Survivors of financial abuse are resistant to the idea of their partner being the nominated payee under UC.^{[footnote 246]} Claimants may seek to avoid joint assessment, either through choosing to live separately from a partner or by failing to disclose a coresident partner, because they wish to avoid situations that could expose them (or their children) to abuse. This point is speculative, however, and it was not addressed in any of the identified sources or interviews. The identified literature does show, however, that financial abuse is often a factor contributing to relationship breakdown, including for couples undergoing joint assessment.^{[footnote 247] [footnote 248]}

Financial abuse is one of the situations in which APAs for UC can be used, including splitting the benefit payment between two bank accounts. However, concerns have been raised that people experiencing financial or domestic abuse may be reluctant to share their experiences with officials, in part because of the risk of the abuser retaliating after becoming aware of the disclosure.^{[footnote 249]} However, this evidence review did not identify any empirical research on APAs and how their use may be shaped by relationship dynamics and abuse.

The process of being assessed as LTAMC may deter some claimants from undergoing joint assessment

Other factors that may deter claimants from moving in with a partner (or declaring a coresident partner) include a desire to avoid the bureaucratic implications of registering a change of status.^{[footnote 250] [footnote 251]} In Koch’s ethnographic study of a council estate in England, claimants resented the time and effort required for registering a new partner with the authorities, especially if their partner was perceived to be unreliable or the arrangement was likely to be short-lived.^{[footnote 252]} One interviewee suggested that being assessed for the purposes of LTAMC may be perceived to be intrusive by claimants, although this was not mentioned in the identified literature.

One interviewee expressed concerns that the assessment process may be “a bit subjective at the edges … the boundaries of those categories [e.g. couple, household, LTAMC] might be a bit fluid”. Another interviewee talked about experiences of their research participants where it was deemed that they had committed living together fraud when actually their partner was not a resident, but did not have another address to use for official purposes. However, the possibility that claimants might perceive the assessment process as subjective or unfair (and therefore be reluctant or unwilling to engage with it) was not raised in any of the identified literature sources. One interviewee also commented that it is quite hard to prove that someone is not living a partner.

Discussion

Few of the studies cited in this chapter relate directly to LTFE, but there is some evidence to suggest that the benefits system may discourage some people from living with a partner. The couple rate may create (real or perceived) financial disincentives to living with a partner, and claimants may also have concerns about access to benefit income and how joint assessment will affect relationship dynamics, including the risk of financial and other forms of domestic abuse. In addition to their effect on partnership decisions, these factors may lead some claimants to misrepresent their circumstances to the authorities. There is a need for further research to explore whether these factors contribute to LTFE and, if so, how this might be addressed by living together policy.

Issues described in this chapter may in some respects be more pronounced under UC compared with under the previous, ‘legacy’ system. Some authors noted that the ‘all or nothing’ quality of UC assessment^{[footnote 253]} creates higher stakes for claimants who may be considering a joint claim. As expressed in one source, “In many ways – and even more than under the previous means-tested system – the presence, resources and needs, and actions of one partner affect the other [under UC].”^{[footnote 254]} The issues explored in this chapter are not new. To some extent, they are an inevitable consequence of means-testing at the household level. However, with UC consolidating multiple benefits into a single payment, paid – in the vast majority of cases – to one adult in the household, these issues are more pertinent than ever.

The following chapter considers policy options to address or mitigate these issues, drawing on international examples as well as evidence from the UK.

Options for policy reform

This chapter presents options for policy reform, drawing on evidence from the UK and other OECD countries, as well as expert interviews. There is a lack of empirical research and policy evaluations in this area from which to draw firm conclusions about what ‘works’ (or does not work) in preventing LTFE. Reflecting the complexity of the issue and the lack of policy evaluations in this space, the policy discussion is descriptive rather than directive (i.e. no recommendations are made). However, the chapter does reflect on the advantages and disadvantages of different policy options, in as much as these were discussed in the identified literature and expert interviews.

Box 4.1: Options for policy reform

The main options for policy reform discussed in the literature are:

Move away from the household as the unit of assessment, for instance by placing greater emphasis on benefits and services that are not means-tested or by taking certain elements out of UC and establishing them as separate, individual benefits. This represents radical change, which may not be affordable or desirable from the perspective of targeting resources at households most at risk of poverty (disproportionately lone parents or single adults living alone).

In troduce individual allowances and replace the joint income test with a partner income test. This approach excludes the spouses or partners of high earners from accessing means-tested benefits, without assuming that all income is shared within couple households.

Make split payments the default under UC or widen the circumstances in which these can be used. This would enable access to independent income, which is particularly important in the context of financial abuse and coercive relationships.

Allow greater flexibility in who should be considered a couple for the purposes of joint assessment, for instance by allowing a transition period for new relationships or allowing claimants or decision makers greater input into how relationships are classified. This would make the decision to move in with a new partner less risky for claimants, but flexibility might result in inconsistency.

Move away from the household as the unit of assessment

In term individualisation is sometimes used to refer to a reform agenda in which social security systems move towards a basis of individual entitlements and away from household means-testing.^{[footnote 255]} As described by Millar,^{[footnote 256]} full individualisation describes a benefit system in which:

• each person has an individual right to claim; no one can apply in the capacity of an adult dependent of another claimant
• assessments of financial need would take place on an individual basis, without taking into account the needs or resources of other adults in the family or household
• the award would cover the needs of that individual only and would not include any payments for adult ‘dependants’
• payments would be made to the individual, so that each individual adult would receive money in their own right

Elements of the UK benefits system are ‘individualised’, since claims are made on an individual basis and cannot be made on behalf of other dependent adults. In other respects, the UK system does not confirm to the individualisation model, since benefit eligibility and entitlement is set according to household (rather than individual) needs and resources, and under UC the default is for payment to be made to one adult on behalf of the household. Taking steps towards individualisation was suggested by several interviewees as a means to address LTFE. In the words of one expert interviewed:

It’s not necessarily about how you operationalise the definition of the household; it’s the fact that you are looking at the household in the first place that introduces a lot of these quite intractable issues”.

Living together fraud and error would cease to exist in a fully individualised welfare system

A fully individualised benefits system does not incentivise or reward certain household types, removing disincentives to co-residential partnering that may be created by the current system of household means-testing.^{[footnote 257]} From the same perspective, individualising the benefits system removes any potential incentive not to declare a partner to the authorities. Put simply, LTFE ceases to exist when the household is no longer the unit of assessment. Full individualisation avoids concerns about disparities in treatment between married and unmarried (or civil partnered) couples, since relationship status is no longer a factor taken into consideration.^{[footnote 258]} Individualisation has been advocated for on the grounds of gender equality, since it gives women access to independent income and improves work incentives for second earners.^{[footnote 259] [footnote 260]}

Despite the advantages of full individualisation, this option is generally not considered viable or realistic.^{[footnote 261] [footnote 262]} A key argument against full individualisation is the heavy financial cost it would incur,^{[footnote 263]} since a larger proportion of the population would be eligible for means-tested welfare. However, one interviewee argued that certain costs associated with the current system – such as additional public money spent on housing supporting two households rather than one – are rarely acknowledged or factored into decision making. In addition to the financial impact, full individualisation has been rejected on the grounds that it would direct resources to low-earning adults who enjoy a good standard of living because of their household circumstances.^{[footnote 264]} However, one interviewee expressed scepticism about whether this group would claim individualised benefits in large numbers if there were additional requirements (e.g. work conditions imposed) associated with doing so. Full individualisation was widely expected by interviewees to be rejected by policymakers.

Steps may be taken towards individualisation without abandoning the household as the unit of assessment

Although full individualisation is generally not regarded as a realistic prospect, there could be scope to take steps towards individualisation within the UK benefits system.^{[footnote 265]} Partial individualisation can take different forms, but broadly speaking it refers to policy adjustments that individualise certain elements of the welfare system or reduce reliance on household means-testing. Although not clearly defined in the literature, partial individualisation might relate to a range of policy reforms:

• Benefit structure and design. This would involve removing benefit income intended for certain purposes from joint assessment and directing it to the individual in a way that is not tied to their partner’s income or resources.
• Whether claims are made on an individual or household basis. This would involve ensuring that claims are made on an individual basis; not allowing claims to made on behalf of other adults classed as dependent.
• Assessment and means-testing. In the most extreme form, this would involve not taking into consideration the income or resources of a partner; under more moderate reforms, the income or resources of a partner might be taken into consideration in a partial or restricted way, for instance only after individual allowances are taken into account (see Box 4.2 for an example of this from Australia).
• Payment of benefits. Even when income is jointly assessed, payment is made to the individual.

As described by one source, “some combination of an individual- and family-based system may be the best way to recognise both independence and interdependence in people’s lives”.^{[footnote 266]} One interviewee argued that, to address LTFE, “the relative importance of living as a couple should be downgraded in our benefits system”.

Box 4.2: International example: Australia

Sharing with the UK a strong emphasis on means-testing,^{[footnote 267] [footnote 268]} Australia represents an example of partial individualism in the welfare system.^{[footnote 269] [footnote 270]} Policy reforms introduced in Australia in 1995 abolished the joint income test for couples. Entitlement to benefits was established on an individual basis, with eligibility based on being a jobseeker (available for and seeking work), a parent (caring for children aged under 16) or a partner (this latter category was introduced to offer protection for women born before 1995 who lacked recent employment experience).^{[footnote 271]} To counteract the high economic cost of an individualised benefits system,^{[footnote 272]} the Australian government replaced the joint income test with a partner income test (discussed below),^{[footnote 273]} meaning that the spouses or partners of high earners would not be eligible for social assistance.

Rather than a joint income test, the Australian system has an individual (claimant) income test and a partner income test for couples. An income test applies to a number of social security benefits in Australia: Jobseeker Payment (JSP), Youth Allowance (YA), Austudy payment and Parenting Payment (Partnered) (PPP).^{[footnote 274]} The income test is applied sequentially, first for the individual claimant and then for their partner.^{[footnote 275]} The individual’s ordinary fortnightly income^{[footnote 276]} is assessed to see whether it falls below the ‘free area’ of AUS $150 for JSP, PPP and YA job seeker or AUS $437 for YA students or Austudy recipients, i.e. the area that is discounted from the income test.^{[footnote 277]} Any income above the ‘free area’ reduces the benefit amount. For example, for JSP there is a reduction of 50 cents for every dollar for income between AUS $150 and AUS $256 and a reduction of 60 cents for every dollar for income over AUS $256 (for single parents the benefit reduction is set at a flat rate of 40 cents for every dollar for income over AUS $150).^{[footnote 278]} There is a cap whereby the individual is not entitled to any support if their fortnightly income or assets^{[footnote 279]} exceed a certain level.^{[footnote 280]} The threshold applied depends on whether the claimant is single or partnered, whether they live with dependent children and (for assets) whether they are a homeowner.^{[footnote 281]} For claimants living with a partner (and whose income and assets do not exceed these thresholds), a partner income test is also applied to the remaining benefit. For JSP, for instance, the benefit amount will be reduced by 60 cents for every dollar of income the partner has beyond the ‘free area’ of AUS $1,157 per fortnight.^{[footnote 282]} Payment for benefits is individualised in Australia, i.e. both partners receive their own benefit payment.^{[footnote 283]} These policy reforms resulted in a significant transfer of benefit payments from men to women.^{[footnote 284]}

Partial individualisation in the Australian benefits system was not introduced with the explicit aim of reducing LTFE. The primary intention of the 1995 reforms was to strengthen work incentives by reducing high effective marginal tax rates.^{[footnote 285]} The purpose of the reforms was to encourage unemployed workers to take up part-time, casual or temporary work, as well as to stimulate inactive partners or unemployed workers to join the labour market.^{[footnote 286] [footnote 287]} There is some evidence to suggest that the reforms had a positive effect on employment, with an increase seen in the number of social assistance recipients in paid work.^{[footnote 288] [footnote 289]} It was not possible to identify any evidence as part of this evidence review that directly addresses if and how these reforms affected LTFE.

It was not possible to identify any estimates of the prevalence of LTFE in Australia and how this has changed over time. The Australian government does not provide estimates of welfare fraud but does report on detected errors and fraud prosecutions.^{[footnote 290]} Although ‘undeclared family relationships’ is identified as a key area of welfare fraud in Australia,^{[footnote 291]} the government does not publish figures for the number of prosecutions relating to living together fraud. The overall number of prosecutions for welfare fraud in Australia has fallen sharply since 2010.^{[footnote 292]} However, this may reflect a less punitive approach to welfare fraud – Wilcock describes a shift away from a ‘prosecution first’ mentality^{[footnote 293]} – rather than a decrease in fraud per se. Only the strongest cases are pursued, and the vast majority of prosecutions result in convictions.^{[footnote 294]}

Since a partner’s income is still taken into consider in determining benefit level, the complex question of how to define the couple still applies in Australia. In the Australian system, a person is a member of a couple if they have a relationship with another person, they are committed to each other on a permanent or indefinite basis and are either married, in a registered relationship or in a ‘de facto relationship’.^{[footnote 295]} In determining whether two people are in a couple relationship, the following factors are considered (not all factors need to be present):^{[footnote 296]}

• the financial aspects of the relationship
• the nature of the household
• the social aspects of the relationship
• any sexual relationship between the people
• the nature of the people’s commitment to each other

An exception is made in instances where there is known to be domestic violence, in which case two people can be considered not to be a couple regardless of the factors outlined above.^{[footnote 297]} Both partners may be interviewed, asked to confirm and asked to provide additional information.^{[footnote 298]} In some cases, a referee (who cannot be a family member or an ex-partner) may be asked to verify the claimant’s relationship status.^{[footnote 299]}

While there was considerable support for partial individualisation among interviewees, the way in which this principle should be implemented or applied in practice was felt to be a complex issue. Several interviewees suggested that the UK might look to Australia for policy lessons in this regard, since Australia’s approach – see Box 4.2 for more details – factors in a partner’s income while allowing the individual a personalised allowance. Some interviewees argued for building up contributory (social insurance) benefits to reduce reliance on means-tested welfare (social assistance). Another interviewee saw scope to separate out certain elements of support from UC, for instance benefit income intended for children or job seekers, redesigning these as an individual rather than household benefit.

Widen the circumstances in which payment can be split

Making split payments the default under UC – or widening the circumstances in which they can be used – is another option for policy reform. As described earlier in this chapter, one element of individualisation is the payment unit^{[footnote 300]} (which may not be the same as the resource unit, i.e. the unit whose assets and resources are taken into consideration in assessing eligibility for means-tested welfare). Even if the household is the resource unit, the payment unit may be the individual. This is the case under UC, although the individual is paid on behalf of the household, leading to concerns around independent income and financial abuse described in Chapter 3. One policy option widely discussed is splitting UC payments by default or making it easier for claimants to choose to do so.

Split payments would mitigate some of the issues described in this report, such as unequal access to income from benefits.^{[footnote 301] [footnote 302]} If split payments were the default, this would reduce the potential risks associated with domestic abuse survivors requesting APAs.^{[footnote 303]} Introducing new terminology – for instance, ‘separate payments’ rather than ‘split payments’ – could help to distinguish a new system, in which split payments are the default or widely used, from the existing system, in which they are an APA to be considered in exceptional circumstances (abuse or financial mismanagement).^{[footnote 304]}

While single payment is the default in England and Wales, a different approach is taken in some of the devolved nations. In Northern Ireland, couples can choose whether to receive UC into a single account or split equally between two bank accounts. However, take-up of split payments in Northern Ireland has been extremely low.^{[footnote 305]} A consultation by the Scottish government found broad support for split payments; three quarters (76%) of responses to the consultation^{[footnote 306]} supported making this the default.^{[footnote 307]} Following the Social Security (Scotland) Bill 2018, split payments are the default, but claimants can opt for a single payment if they prefer.^{[footnote 308]} Qualitative research suggests that some UC claimants in England and Wales would prefer a split payment system.^{[footnote 309]}

If a split payment system becomes the norm (or a more widely used option), this raises questions about how the award should be split across the two partners.^{[footnote 310]} There are different approaches available to splitting payments, and each has advantages and disadvantages:

• Using a percentage approach: This can be either a 50/50 split or weighted towards one partner. An equal split sends a clear message about equality but may be less appropriate in circumstances where the resources and responsibilities of the two partners differ;^{[footnote 311]} this approach may also raise difficulties in situations where, for instance, one partner’s earnings increase and the other partner’s benefit decreases as a result.
• Directing certain elements – e.g. the child element – to one partner (the lead carer): This would help to ensure that payments intended for children are spent on them.^{[footnote 312]} However, not all couples are comfortable with the notion of a lead carer, viewing themselves as equal co-parents.^{[footnote 313]} This principle could be extended to other elements of UC, for instance payments related to limited work capacity.
• Allowing the couple to decide for themselves how to split the payment: Allowing claimants to decide on the payment approach creates greater flexibility in adapting to individual preferences and circumstances but could be problematic in the context of abusive or coercive relationships.

Among UC claimants who preferred split payments, views differed on the best way to approach this.^{[footnote 314]} Some people felt that equal payments for both partners would be best, whereas others – most commonly women with children – felt that this would disadvantage the partner who took on a greater share of the caring responsibilities.^{[footnote 315]}

Allow greater flexibility in deciding who should be considered a couple household

If the household is retained as the unit of assessment and payment continues to be made to one partner on behalf of the household under UC, then scope for policy reform is more limited. However, some suggestions for reform relate to the process of deciding whether or not two people are in a relationship (and therefore subject to joint assessment), including introducing a transition period, trusting claimants to define their own relationship status and allowing decision makers greater discretion.

A transition period might help couples to decide if they want to undergo joint assessment together

Introducing a transition period (for instance, 6 to 12 months) where couples are not obligated to undergo joint assessment could help to mitigate the ‘all or nothing’ quality^{[footnote 316]} to joint assessment for UC.^{[footnote 317] [footnote 318]} A transition period would enable couples to make a more informed decision about whether they want to form a joint household. However, wider issues of joint assessment, such as lack of access to independent income, would be deferred rather than circumvented.^{[footnote 319] [footnote 320]} A similar approach has been suggested in relation to OECD countries, such as New Zealand (see Box 4.3), facing similar challenges to the UK.

Box 4.3: International example: New Zealand

There are parallels between UK and New Zealand benefit systems in terms of their treatment of partnered and unpartnered claimants. Living together fraud (LTF) accounts for around one third of prosecutions for welfare fraud in New Zealand.^{[footnote 321]} as in the UK, the welfare system in New Zealand relies heavily on means-testing at the household level.^{[footnote 322] [footnote 323]}) In New Zealand, means-tested benefits (e.g. Supported Living Payment, Sole Parent Support) are paid at a higher rate for single adults with children compared with those who are married, in a civil union or in ‘a relationship in the nature of marriage’.^{[footnote 324]} This reflects the economies of scale associated with living with a partner, although commentators have pointed out that this also applies to other forms of co-residence.^{[footnote 325]} A couple rate also applies to benefits in New Zealand, but only a small proportion of benefit claimants in New Zealand (<10%) are from couple households.

It has been suggested that the benefits system in New Zealand relies on outdated assumptions about financial dependence in relationships, assuming that income from benefits is shared (equally) when evidence suggests this is frequently not the case.^{[footnote 326] [footnote 327] [footnote 328]} An assumption is made that a new partner will take responsibility for caring and providing for dependent children in the household, which may not align with preferences and behaviour.^{[footnote 329] [footnote 330]}

The government in New Zealand takes into consideration a range of factors when deciding whether two people are in a relationship: the level of commitment in a relationship and the presence of financial dependence, as well as other factors, such as living together at the same address or frequently staying overnight, shared responsibilities (including childcare), shared bank accounts or credit cards, joint household bills, emotional support and public recognition of the relationship.^{[footnote 331]} Critics have described this as “an arbitrary checklist”,^{[footnote 332]} resulting in a lack of clarity about how relationships will be assessed and evaluated.^{[footnote 333] [footnote 334]} This lack of clarity about decision making may contribute to LTFE.^{[footnote 335]} There is also a certain amount of subjectivity involved in making assessments about relationship status.^{[footnote 336]}

It has been argued that a clearer definition of the distinction between partnered and single people is needed, enabling more effective and consistent decision making from government agencies and offering claimants greater clarity on the process.^{[footnote 337]} In 2013, a ‘relationship status verification’ procedure was introduced, requiring claimants to ask another adult (non-family member) to verify that they are not^{[footnote 338]} in a relationship.^{[footnote 339]} There is a lack of research on the impact of this policy and the impact it has on claimants (or potential claimants). Accusations of LTF are generally made by concerned neighbours or acquaintances, but only a small proportion (2 to 5%) result in prosecutions.^{[footnote 340]} The process of investigating LTF in New Zealand has been described as “protracted and intimidating”.^{[footnote 341]} Penalties (repayments or imprisonment) are harsh compared with such offences as tax evasion,^{[footnote 342] [footnote 343]} and this has implications for child welfare and child poverty.^{[footnote 344]}

Commentators have argued that the welfare system in New Zealand discourages (re)partnering,^{[footnote 345] [footnote 346] [footnote 347]} echoing arguments made about the benefits system in the UK (see Chapter 3). Benefit claimants may avoid entering a new relationship while claiming benefits for fear of prosecution and imprisonment.^{[footnote 348]} With parallels to the UK, claimants who are in unstable or unsupportive relationships face challenges, as do those with dependent children as they may fear that relaying on a partner rather than the state will provide a less generous or stable income for their family.^{[footnote 349]}

Reforms introduced in the Social Security (Fraud Measures and Debt Recovery) Amendment Act 2014 mean that the spouse or partner of someone convicted of LTF is themselves liable for prosecution or debt repayment if they knowingly benefited from the fraud or if they ‘ought to have known’ they were benefiting.^{[footnote 350]} These reforms were designed to enable fairer and more equitable accountability for LTF and to facilitate debt recovery.^{[footnote 351]} Commentators have argued that these reforms are unduly harsh because partners are not held responsible for other financial offences, such as tax avoidance.^{[footnote 352]}

In response to these issues, some commentators have suggested removing or reducing reliance on the couple-based unit of assessment, moving towards greater individualisation in the benefits system.^{[footnote 353] [footnote 354]} It has been argued that it would be possible to completely individualise the welfare system in New Zealand, removing the need for any kind of relationship test.^{[footnote 355]} However, this would be a highly costly option.^{[footnote 356]} Assuming household-level means-testing of welfare is retained, other suggestions include individualising certain elements of the welfare system (i.e. disassociating eligibility for certain benefits from a partner’s income),^{[footnote 357]} placing greater emphasis on co-residence,^{[footnote 358]} allowing claimants to self-define when their relationship is ‘in the nature of marriage’ ^{[footnote 359]} and introducing a transition period where claimants can retain their benefit for a period of time after entering a new relationship.^{[footnote 360] [footnote 361]}

Moving away from a ‘checklist’ approach could allow for greater nuance and diversity in what it means to be a couple

The process of assessing who is (and is not) in a couple relationship could be adjusted to move away from a ‘checklist’ approach. At the most extreme end, some have argued that the government should allow claimants to define for themselves whether they are part of a couple or living in a shared household. ^{[footnote 362]} In the view of some commentators, the government should have more trust in people to share truthful information about their circumstances.^{[footnote 363]} One interviewee described the government’s lack of trust in people to tell the truth about their living circumstances as “a huge underestimation of people, of the public”. Another interviewee observed that placing greater trust in claimants might be mutually reinforcing, i.e. claimants may respond with greater trust and honesty:

I think it would be quite a useful thing to do [placing greater trust in claimants] and I think it would send quite a good signal to the person concerned, or the couple concerned, that you were really trying to be helpful and give them, and I think it would be trust, being forwarded to them, and they hopefully would respond and would forward trust and honestly back to you.

Allowing decision makers greater discretion could allow LTAMC assessments to take into consideration a broader range of factors. As argued by some sources^{[footnote 364]} and interviewees, allowing decision makers greater discretion would allow them to get to know an individual or couple better and to develop an in-depth understanding of their situation. This could reduce the likelihood of couples being misclassified, as well as enabling decision makers to be more responsive to such factors as the risk of domestic violence. As one interviewee explained:

I think one thing that’s kind of stuck with me from looking at different income support systems is that [on one hand] you can have an approach that’s quite bureaucratised, which is the UK approach. So we have a set of rules and we try and apply them in the same way to everyone. And [on the other hand] you can have something that’s closer maybe to some of the Nordic countries where they have much more discretion, right? The manager gets to know the person much better, has fewer people to deal with. They try to understand their situation and … the goal is not just to give them some money, but to try and help them achieve some kind of financial independence, try to help them sustain themselves financially. What works for one person might not necessarily work for another”.

However, discretion in the welfare system is a divisive issue, and other interviewees argued against it. A system with greater discretion is more costly to implement and leaves greater room for bias and discrimination. In the words of one expert interviewed:

It’s too easy for discretion to actually be misused or become racist or become personalised or whatever, so you happen to strike someone who’s pretty strict on exercising the discretion or you happen to look like the ‘wrong sort of person’, or you don’t appear polite enough or whatever and you pay the price.

Discussion

There are compelling arguments for greater individualisation in the benefits system, not least because the less reliant the benefit system is on the household as the unit of assessment, the more limited the scope for LTFE. However, focusing on the household rather than the individual enables resources to be directed to those most at risk of experiencing financial hardship, and poverty is disproportionately concentrated in households comprised of an adult living without a partner (with or without dependent children). Between the two extremes of a fully individualised and household-based approach lies a grey area comprised of a multitude of possible policy arrangements. Elements of individualisation may be introduced without abandoning the household as the unit of assessment, for instance by introducing individual allowances or making the individual the default payment unit for UC.

Defining what it means to be a couple or to live in a joint household is inherently complex, and some have argued for greater flexibility in how the LTAMC rule is applied. The application of the LTAMC rule in practice will inevitably assume financial support where there is none, and vice versa. There have been calls for more flexibility in decision making, allowing claimants to self-define their relationship status or giving decision makers greater discretion to take the individual’s circumstances into account. However, these approaches are not without their limitations. Individuals may not be truthful about their circumstances (particularly if they stand to lose out financially), and greater discretion in decision making increases the risk of bias and discrimination. Another option for applying the LTAMC rule is to introduce a transition period for couples in new relationships, i.e. a period of a few months during which they can opt out of joint assessment. While a transition period delays rather than circumvents most of the issues associated with household means-testing, it enables claimants to test out the strength of the relationship and the reliability of their partner, allowing them to make a more informed choice about undergoing joint assessment.

This is a complex policy area, with no clear or easy solutions. Options for reform discussed in this chapter are all associated with advantages, disadvantages and trade-offs. There is a lack of empirical research and policy evaluations in this area, and this limits our understanding of what works (or does not work) in preventing LTFE.

Conclusions

Drawing together the evidence on living together policy and LTFE for the first time, this report highlights limitations associated with the evidence base on this topic:

• Few research studies directly address or refer to LTFE, whether in the UK or in other OECD countries. Despite the lack of direct evidence, the reviewed literature points to certain mechanisms that may plausibly affect or contribute to LTFE, although it is important to note that in some cases these mechanisms have not be empirically tested.
• Literature in this area is predominantly comprised of small-scale, qualitative studies. While these studies generate rich data and valuable insights, no impact assessments of living together policy were identified from which to draw causal inferences. Conclusions about the effects of living together policy on claimant experiences and behaviour should therefore be taken as indicative rather than definitive.
• In some areas, links between living together policy and LTFE are theorised but not empirically tested or substantiated. For instance, there is a need for further research to explore whether joint assessment or single payment for UC contributes to financial and other forms of domestic abuse.

Despite these limitations, the available literature provides some insight into the social and political context in which LTFE is embedded and points to how policy change might be levered to create a welfare system where the risk of LTFE is lower. There is, however, a need for further research to substantiate the extent to which these factors have an effect on LTFE, as well as to evaluate the effect of policy reforms introduced in the UK and other OECD countries to address the issue.

The increasing instability and plurality of family forms may make it more complex to delineate the couple and identify the benefit unit. A growing number of people are rejecting marriage or delaying it until later in life. Cohabitation is on the rise, whether as an alternative or a precursor to marriage. Cohabiting partnerships are on average more unstable than marriages, although this may reflect differences in the characteristics of people who chose cohabitation relative to marriage. A growing number of couples are living apart together (i.e. in separate households), for a variety of reasons. Increasingly, society is comprised of a diversity of family forms: single-parent families, blended families, extended families, etc. It was not possible to identify any evidence to link these sociodemographic changes to LTFE. However, these changes may make the application of living together policy more complex, and they may call some of the assumptions on which it is based (for instance, about financial support) into question.

The configuration of the UK benefits system may discourage some people from living with a partner and could theoretically deter people from declaring a partner to the authorities. The couple rate may deter some claimants from living with a partner, particularly if it expected that they will be worse off financially. People may also avoid joint assessment with a partner as a means of protecting their financial independence and autonomy. Research has highlighted the importance of financial independence to people on low incomes, particularly women. Claimants may not have (full) access to jointly assessed income from benefits, particularly if it is paid into a single bank account, as is the default under UC. These concerns are particularly important for claimants who have reasons to doubt the reliability of their partner – for instance if their partner is over-indebted; has a history of insecure employment; or has issues with drugs, alcohol or gambling – or who are in a new relationship. Undergoing joint assessment with a partner may be risky from the perspective of financial and other forms of domestic abuse, although further research is needed to substantiate this.

This is a complex policy area, with no clear or easy solutions. Different approaches are associated with advantages, disadvantages and trade-offs. LTFE ceases to exist if the household is no longer the unit of assessment. However, a fully ‘individualised’ benefits system would be costly and less effective at targeting resources at those most at risk of experiencing financial hardship. Steps may be taken towards individualisation without abandoning the household as the unit of assessment. In this regard, the UK might look to Australia’s partially individualised welfare system for policy lessons. The Australian system relies on a partner income test rather than a couple income test – an individual allowance is applied before the partner’s income is taken into consideration. Another form of individualisation would be to make split payments the default under UC or widen the circumstances in which these can be applied. However, this raises complex questions about how the award should be split across the two partners (50/50, reflecting roles and responsibilities or allowing couples to decide). The impact of the LTAMC rule on LTFE might be mitigated by allowing couples greater freedom, whether by introducing a transition period in which couples do not have to undergo joint assessment or by allowing claimants the autonomy to define their own relationship status.

Annex 1: Search protocol

This document outlines the databases, search terms, criteria and procedures used in Evidence Review 4 (ER4).

Objective of the review

The aim of this review (ER4) was to gather evidence on the UK’s LTAMC policy,^{[footnote 365]} specifically: (a) how the LTAMC policy contributes to welfare fraud and error; (b) how this relates to contextual factors, such as changing social attitudes, gender relations and patterns of household formation; and (c) how the LTAMC policy compares with international examples from other OECD countries.

Box A1.1: Research questions

1. What are the prevailing attitudes and approaches to household formation in the UK and internationally (OECD countries)?

• Do these attitudes impact on the dynamic cycle of household formation and break up?
• Do these approaches impact welfare fraud and error rates?

2. How can a household or benefit unit be defined in social policy?

• What is the criteria for a stable or unstable household unit in the UK?
• What type of households are re-partnering in the UK?
• How does this compare with international policy (OECD countries)? Does this impact on welfare fraud and error rates?

3. What is the relationship between household composition and household finances?

• Are these shaped by socio-economic or geographical or age factors?
• How has this changed over time?
• How does this compare internationally (OECD countries)? Is this reflected in social policy?

4. What is the role of women in the household?

• Do women want, need or seek financial independence from male partners?
• Has this changed over time, and is this represented in UK and international social policy (OECD countries)?
• How does this impact welfare fraud and error rates?

5. Does the living together policy affect partnering and cohabitation decisions? If so, how?

Overview of the approach

Two QSRs were conducted:

Evidence Review 4A covered patterns in household formation and the relationship between household type and financial status: RQ1, RQ1a, RQ3, RQ3a, RQ3b, RQ3c, RQ4, RQ4a, RQ4b, RQ5.

Evidence Review 4B covered how households are defined for the purposes of social policy and the implication of this for benefit fraud and error: RQ1b, RQ2, RQ2a, RQ2b, RQ2c, RQ3c, RQ4b, RQ4c, RQ5.

Steps followed as part of this review

Step 1: Identify and refine databases, search terms and inclusion and exclusion criteria

We carried out a targeted search of key data sources (Evidence Review 4A), as well as a structured search of Web of Science and Google Scholar (screening the first 30 results for each database)^{[footnote 366]} to identify relevant academic literature (Evidence Review 4A and Evidence Review 4B).

Targeted search

For Evidence Review 4A, we conducted a targeted search of key data sources (Box A1.2) to gather data on patterns and trends pertinent to the search questions.

Box A1.2: Data sources

• Attitudes to household formulation
  • British Social Attitudes Survey (UK)^{[footnote 367]}: attitudes to marriage
  • International Social Survey Programme (international)^{[footnote 368]}: attitudes to family
• Patterns of household formation
  • Eurostat (EU + UK): distribution of households by household type^{[footnote 369]}
• Household composition and household finances
  • Eurostat (EU + UK): distribution of households by household type and income
• The role of women in the household
  • British Social Attitudes Survey (UK): attitudes to gender roles and women’s work outside the home
  • ISSP (international): attitudes to gender roles

Structured search

The structured search for Evidence Review 4A and Evidence Review 4B used a Boolean search string to ensure the relevance of results (see Table A1.1). We conducted four separate searches (see Table A1.1), repeating each twice, once for the UK and once for other OECD countries.^{[footnote 370]} We drew upon the search strings suggested in our proposal, refined in consultation with the DWP and our expert team, and tested we these strings for relevance.

Table A1.1: Search terms that we used in our search (Google Scholar and Web of Science)

Web of Science and Google Scholar					Web of Science	Google Scholar
ER4A
Patterns of household formation	(“household type” OR “household composition” OR “lone parent” OR “stable relationship” OR “marriage” OR “partnership decision*”) AND (“patterns OR “trends” OR “changes”)	AND	(“UK”)	OR	(“Australia” OR “Austria” OR “Belgium” OR “Canada” OR “Chile” OR “Colombia” OR “Costa Rica” OR “Czech” OR “Denmark” OR “Estonia” OR “Finland” OR “France” OR “Germany” OR “Greece” OR “Hungary” OR “Iceland” OR “Ireland” OR “Israel” OR “Italy” OR “Japan” OR “Korea” OR “Latvia” OR “Lithuania” OR “Luxembourg” OR “Mexico” OR “Netherlands” OR “New Zealand” OR “Norway” OR “Poland” OR “Slovak” OR “Slovenia” OR “Spain” OR “Sweden” OR “Turkey” OR “United States”)	(“EU” or “Europe” or “OECD”)
Household type and finances	(“household composition” OR “household type” “lone parent”) AND (“income” OR “poverty” OR “deprivation” OR “low income”)	AND	(“UK”)	OR	(“Australia” OR “Austria” OR “Belgium” OR “Canada” OR “Chile” OR “Colombia” OR “Costa Rica” OR “Czech” OR “Denmark” OR “Estonia” OR “Finland” OR “France” OR “Germany” OR “Greece” OR “Hungary” OR “Iceland” OR “Ireland” OR “Israel” OR “Italy” OR “Japan” OR “Korea” OR “Latvia” OR “Lithuania” OR “Luxembourg” OR “Mexico” OR “Netherlands” OR “New Zealand” OR “Norway” OR “Poland” OR “Slovak” OR “Slovenia” OR “Spain” OR “Sweden” OR “Turkey” OR “United States”)	(“EU” or “Europe” or “OECD”)
ER4B
Household as a benefit unit	(“household target*” OR “benefit unit”)	AND	(“UK”)	OR	(“Australia” OR “Austria” OR “Belgium” OR “Canada” OR “Chile” OR “Colombia” OR “Costa Rica” OR “Czech” OR “Denmark” OR “Estonia” OR “Finland” OR “France” OR “Germany” OR “Greece” OR “Hungary” OR “Iceland” OR “Ireland” OR “Israel” OR “Italy” OR “Japan” OR “Korea” OR “Latvia” OR “Lithuania” OR “Luxembourg” OR “Mexico” OR “Netherlands” OR “New Zealand” OR “Norway” OR “Poland” OR “Slovak” OR “Slovenia” OR “Spain” OR “Sweden” OR “Turkey” OR “United States”)	(“EU” or “Europe” or “OECD”)
Welfare fraud and error	(“benefit fraud” AND “living together” OR “cohabit*)	AND	(“UK”)	OR	(“Australia” OR “Austria” OR “Belgium” OR “Canada” OR “Chile” OR “Colombia” OR “Costa Rica” OR “Czech” OR “Denmark” OR “Estonia” OR “Finland” OR “France” OR “Germany” OR “Greece” OR “Hungary” OR “Iceland” OR “Ireland” OR “Israel” OR “Italy” OR “Japan” OR “Korea” OR “Latvia” OR “Lithuania” OR “Luxembourg” OR “Mexico” OR “Netherlands” OR “New Zealand” OR “Norway” OR “Poland” OR “Slovak” OR “Slovenia” OR “Spain” OR “Sweden” OR “Turkey” OR “United States”)	(“EU” or “Europe” or “OECD”)

Note: All searches were restricted to sources published between 2011 and 2021; geographical indicators differ because Google Scholar allows fewer search terms than Web of Science.

Step 2: Carry out the search

We conducted the full search (using the above parameters) and collected all sources (first 30 sources for each search).

Step 3: Remove duplicates, screen and select sources for the review

We removed duplicates that resulted from searches in multiple databases.

We screened titles and abstracts against inclusion or exclusion criteria (see Table A1.2) to confirm whether each source would be selected for full-text review.

Literature was screened by one researcher, with a sample screened also by a second researcher to ensure that inclusion or exclusion criteria were applied consistently. Up to 50 sources were included following screening.

Table A1.2: Inclusion and exclusion criteria

Inclusion criteria	Exclusion criteria
Academic articles and grey literature	Commentaries, editorials and features
Articles published in 2011 or later	Articles published before 2011
Articles relating to the UK or other OECD country	Articles publishes outside both the UK and the OECD
Articles published in English	Articles published in languages other than English

Step 4: Review full texts

We reviewed the full text of sources identified from the previous steps. We first reassessed the full text against inclusion or exclusion criteria to ensure their relevance and then extracted information relevant to the research questions.

To structure each review, we developed a data extraction tool (Table A1.3) to record information from the reviewed sources, including capturing the details of the source and the data needed to respond to research questions.

Table A1.3: Data extraction tool

Citation information
Full reference
URL
Authors
Publication date
Abstract
Publication type (book, journal article, report, survey etc.)
Geographic scope

Research method
Methodology type (e.g. quantitative – primary data collection)
Methodology detail
Main outcome or indicator

Data quality
Validity	Does the indicator capture the phenomenon we are interested in? – note any concerns
Generalisability	Are the findings restricted to a specific population group? – note any limitations
Reliability	Is there anything to suggest the results would not be replicated if the research were reproduced under the same conditions? – note any concerns
Transparency	Is the method clear, including assumptions made? – note any concerns
Overall	Overall quality assessment (low, medium, high)

Findings
Household formation	Attitudes to household formation and gender roles within the household
Household formation	Patterns or trends in household formation
Household formation	Factors affecting decisions about household formation
Household formation	Relationship between household composition and finances (and mediating factors, e.g. socio-economic or geographical or age)
Living together fraud and error	Prevalence of LTFE (including change over time)
Living together fraud and error	Variation across groups (including household types) and risk factors for living together fraud and error
Living together fraud and error	Factors contributing to LTFE
Living together fraud and error	Policy changes recommended (reducing LTFE)
Defining the household for the purposes of social policy	Name of policy
Defining the household for the purposes of social policy	Definition of household or benefit unit
Defining the household for the purposes of social policy	Assessment criteria and decision-making process
Defining the household for the purposes of social policy	Debates or controversies
Defining the household for the purposes of social policy	Impact on welfare fraud and error
Defining the household for the purposes of social policy	Effect on household formation
Defining the household for the purposes of social policy	Other outcomes or impact or implication associated with policy
Defining the household for the purposes of social policy	Policy changes recommended (defining the household for the purposes of social policy)

Findings
	How and where was source identified (structured search or snowballing, database)
Citation tracking	List relevant sources to follow up
	Exclude source (Y/N)
	Reason for excluding source
Notes	Anything else relevant to add

We reviewed up to 50 sources in this way.

Step 5: Snowball search

We reviewed the bibliographies of studies identified for inclusion in this study to assess whether they reference other relevant studies (snowballing). We compiled a list of these studies and followed the same steps as detailed above to determine whether full-text review and data extraction were pertinent.

Annex 2: Interview topic guide

Introduction

1. To start us off, it would be great if you could introduce yourself and give a brief overview of your work and research in this area.

UK policy context (LTAMC)

The discussion will focus on how to define the household (or benefit unit) for the purposes of assessing eligibility for household-level benefits. I would first like to focus on definitions of the household or benefit unit in UK social policy.

2. What are the key challenges associated with defining the household or benefit unit in UK social policy?

Prompts

• What are the implications of this?
• To what extent and why are these challenges specific to the UK context?

3. To the best of your knowledge, how does the Department for Work and Pensions (DWP) define living together as a household or benefit unit?

Prompts

• Who makes these decisions?
• What information is used to inform these decisions?
• What criteria are used to assess whether people are in a relationship or part of the same household?

I would now like to discuss some of the strengths and limitations of the LTAMC policy.

4. In your view, what are the main strengths and limitations of the LTAMC policy?

Prompts

• What is working well about the LTAMC policy?
• What is working less well?

5. What assumptions underpin the LTAMC policy? To what extent are these reasonable assumptions and why?

Prompts

• assumptions about relationships: commitment, fidelity or monogamy, stability, predictability, longevity, emotional bond, shared surname, sexual component
• assumptions about financial dependence: sharing resources, taking financial responsibility for one another

6. How well aligned is the LTAMC policy with the reality of how people form domestic partnerships? What (if any) are the areas of discontent or misalignment?

Prompts

• stability: how frequently relationships begin and end
• regularity: how often a couple share a residence and how predictable this is
• definition: how clearly relationships are defined and articulated
• financial arrangements: if and how partners share financial resources and desire independence from one another

7. How, if at all, have broader social and economic changes over time affected the relevance and impact of the LTAMC policy?

Prompts

• patterns of household formation and dissolution
• the relationship between household composition and household finances
• the role of women in the economy and society
• expectations around financial (in)dependence in relationships

Thinking now about the implications and consequences of the Living Together as a Married Couple (LTAMC) policy …

8. What unexpected or unanticipated consequences might the LTAMC policy have?

Prompts

• What are the implications of this?

9. What consequences (if any) does the LTAMC policy have for how people make decisions regarding partnership and co-residence?

Prompts

• forming and formalising a relationship
• sharing a residence all or some of the time
• sharing financial information with a partner
• making joint financial decisions

• dissolving a relationship or leaving a partner

• How (if at all) has this changed over time and why?

10. What consequences (if any) does the LTAMC policy have for women’s equality and independence?

Prompts

• financial (in)dependence
• risk of domestic violence or ability to leave an abusive relationship

• risk of economic abuse

• How (if at all) has this changed over time and why?

11. What consequences (if any) does the LTAMC policy have for fraud and error in the welfare system?

Prompts

• the rate of fraud and error

• risk factors or groups associated with fraud and error

• How (if at all) has this changed over time and why?

12. What changes (if any) would you make to the LTAMC policy and why?

Prompts

• What would be the implications of this change for …?
• the rate of welfare fraud and error
• risk factors or groups affected by welfare fraud and error
• decisions regarding partnership and co-residence
• women’s equality and independence

International policy context

Now thinking about how other OECD countries approach this issue …

[The focus is on other OECD countries, but examples from non-OECD countries can be included if highly pertinent]

13. What different approaches are taken internationally to defining the household or benefit unit for the purposes of social policy?

Prompts

• country or countries

• policy name or description

• How does this policy or approach differ from the UK?

[For each policy or approach mentioned]

14. What are the main strengths and shortcomings of this policy or approach?

Prompts

• What is working well about this policy or approach?

• What is working less well?

• How does this compare to the UK?

15. What assumptions underpin this policy or approach? To what extent are these reasonable assumptions and why?

Prompts

• assumptions about relationships: commitment, fidelity or monogamy, stability, predictability, longevity, emotional bond, shared surname, sexual component

• assumptions about finance dependence: sharing resources, taking financial responsibility for one another

• How does this compare to the UK?

16. What consequences (if any) does this policy or approach have for how people make decisions regarding partnership and co-residence?

Prompts

• forming and formalising a relationship
• sharing a residence all or some of the time
• sharing financial information with a partner
• making joint financial decisions

• dissolving a relationship or leaving a partner

• How does this compare to the UK?

17. What consequences (if any) does this policy or approach have for women’s equality and independence?

Prompts

• financial (in)dependence
• risk of domestic violence or ability to leave an abusive relationship

• risk of economic abuse

• How does this compare to the UK?

18. What consequences (if any) does this policy or approach have for fraud and error in the welfare system?

Prompts

• the rate of fraud and error

• risk factors or groups associated with fraud and error

• How does this compare to the UK?

19. What could the UK apply or learn from this/these approach(es)?

Prompts

• To what extent is this policy or approach (or elements of it) transferable to the UK context and why?

• What implications would it have if the UK took this approach?

Living together fraud and error

We have touched on the implications of defining the household or benefit unit for welfare fraud and error, and I would like to focus on this for the rest of the discussion.

20. How common is living together fraud (and error) in the UK?

Prompts

• To what extent is the true rate of living together fraud (and error) in the UK reflected in official statistics? Is it over- or under-reported and why?
• How (if at all) has the prevalence of living together fraud (and error) in the UK changed over time?

21. How does the rate of living together fraud (and error) in the UK compare to other OECD counties and why?

Prompts

• Which countries perform well in minimising the rate of living together fraud (and error)?

[For countries identified as good performers]

22. What is it about [country name] that reduces or minimises the rate of living together fraud (and error)?

Prompts

• policy factors – definition of the household and assessment criteria, complexity of welfare system, reliance on targeting or means-testing, unit of targeting or means-testing (individual, household), assessment process
• economic factors – inequality, poverty or material deprivation

• social or cultural factors – attitudes to relationships, patterns of household formation and dissolution, prevalence of different household types

• What could the UK learn from [country name]?

23. What are the key drivers contributing to living together fraud (and error) in the UK and internationally?

Prompts

• How well understood are the drivers of living together fraud (and error) by research? And by and policymakers?
• How have the drivers of living together fraud (and error) changed over time and why?

• How does the UK compare to other countries and why?

• How important is the policy context, in particular the definition of the household or benefit unit?

24. To what extent is living together fraud intentional or deliberate, i.e. to what extent does it reflect error rather than fraud?

Prompts

• What makes you say that?
• What are the implications of this?

Conclusion/wrap-up

25. If you could change one thing to reduce the rate of living together fraud (and error) in the UK, what would it be and why?

Prompts

• And if you could change one thing in relation to government policy (LTAMC), what would it be?

26. Is there anything else that you would like to add that we haven’t mentioned so far? 27. Are you able to recommend anyone else with expertise in this area that we might interview as part of this study?

1. UK Government (2021) ‘Fraud and error in the benefit system for financial year ending 2021’ (viewed on 2 March 2022) https://www.gov.uk/government/statistics/fraud-and-error-in-the-benefit-system-financial-year-2020-to-2021-estimates/fraud-and-error-in-the-benefit-system-for-financial-year-ending-2021#def-1 ↩

2. UK Government (2013) ‘Advice for decision makers: staff guide’, Chapter E4, Universal credit – Living Together as a Married Couple, London: Department for Work and Pensions (viewed on 6 May 2022) https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/661551/adme4.pdf ↩

3. National Audit Office (2015) ‘Understanding fraud and error in benefits and tax credits: a primer’, London: Department for Work and Pensions (viewed on 4 May 2022) https://www.nao.org.uk/wp-content/uploads/2015/09/Understanding-fraud-and-error-a-primer.pdf ↩

4. National Audit Office (2015) ‘Understanding fraud and error in benefits and tax credits: a primer’, London: Department for Work and Pensions (viewed on 4 May 2022) https://www.nao.org.uk/wp-content/uploads/2015/09/Understanding-fraud-and-error-a-primer.pdf ↩

5. National Audit Office (2015) ‘Understanding fraud and error in benefits and tax credits: a primer’, London: Department for Work and Pensions (viewed on 4 May 2022) https://www.nao.org.uk/wp-content/uploads/2015/09/Understanding-fraud-and-error-a-primer.pdf ↩

6. UK Government (2013) ‘Advice for decision makers: staff guide’, Chapter E2, Benefit unit, awards and maximum amount, London: Department for Work and Pensions (viewed on 5 May 2022) https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1040207/adme2.pdf ↩

7. Millar, J (2004) ‘Squaring the circle? Means testing and individualisation in the UK and Australia’, Social Policy and Society, 3 (1), 67-74 ↩

8. Department for Work and Pensions (2020) ‘Fraud and error in the benefit system: background information and methodology’, London: Department for Work and Pensions (viewed on 5 May 2022) https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/888425/fraud-and-error-statistics-background-and-methodology-revised-29-may-2020.pdf ↩

9. Gugushvili, D and Hirsch, D (2014) ‘Means-tested and universal approaches to poverty: international evidence and how the UK compares’, working paper 640, Loughborough: Centre for Research in Social Policy ↩

10. Gugushvili, D and Hirsch, D (2014) ‘Means-tested and universal approaches to poverty: international evidence and how the UK compares’, working paper 640, Loughborough: Centre for Research in Social Policy ↩

11. Organisation for Economic Co-operation and Development (viewed on 4 May 2022) https://www.oecd.org/about/members-and-partners/ ↩

12. UK Government (2021) ‘Fraud and error in the benefit system for financial year ending 2021’ (viewed on 2 March 2022) https://www.gov.uk/government/statistics/fraud-and-error-in-the-benefit-system-financial-year-2020-to-2021-estimates/fraud-and-error-in-the-benefit-system-for-financial-year-ending-2021#def-1 ↩

13. National Audit Office (2015) ‘Understanding fraud and error in benefits and tax credits: a primer’, London: Department for Work and Pensions (viewed on 4 May 2022) https://www.nao.org.uk/wp-content/uploads/2015/09/Understanding-fraud-and-error-a-primer.pdf ↩

14. National Audit Office (2015) ‘Understanding fraud and error in benefits and tax credits: a primer’, London: Department for Work and Pensions (viewed on 4 May 2022) https://www.nao.org.uk/wp-content/uploads/2015/09/Understanding-fraud-and-error-a-primer.pdf ↩

15. National Audit Office (2015) ‘Understanding fraud and error in benefits and tax credits: a primer’, London: Department for Work and Pensions (viewed on 4 May 2022) https://www.nao.org.uk/wp-content/uploads/2015/09/Understanding-fraud-and-error-a-primer.pdf ↩

16. Sum of fraud, claimant error and official error. ↩

17. UK Government (2021) ‘Tables: fraud and error in the benefit system, 2020 to 2021 estimates (XLS)’ (viewed on 4 May 2022) https://www.gov.uk/government/statistics/fraud-and-error-in-the-benefit-system-financial-year-2020-to-2021-estimates ↩

18. UK Government (2021) ‘Tables: fraud and error in the benefit system, 2020 to 2021 estimates (XLS)’ (viewed on 4 May 2022) https://www.gov.uk/government/statistics/fraud-and-error-in-the-benefit-system-financial-year-2020-to-2021-estimates ↩

19. UK Government (2021) ‘Tables: fraud and error in the benefit system, 2020 to 2021 estimates (XLS)’ (viewed on 4 May 2022) https://www.gov.uk/government/statistics/fraud-and-error-in-the-benefit-system-financial-year-2020-to-2021-estimates ↩

20. UK Government (2013) ‘Advice for decision makers: staff guide’, Chapter E4, Universal credit – Living Together as a Married Couple’, London: Department for Work and Pensions (viewed on 6 May 2022) https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/661551/adme4.pdf ↩

21. Organisation for Economic Co-operation and Development (viewed on 4 May 2022) https://www.oecd.org/about/members-and-partners/ ↩

22. Collins, AM Coughlin, D Miller, J and Kirk, S (2015) ‘The production of quick scoping reviews and rapid evidence assessments: a how to guide’, London: Department for Environment, Food and Rural Affairs https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/560521/Production_of_quick_scoping_reviews_and_rapid_evidence_assessments.pdf ↩

23. This term refers to using the reference list of one or more selected papers to identify additional papers that are relevant to the review. ↩

24. Due to technical issues, one interview was not recorded. In this case, analysis was based on interview notes. ↩

25. Collins, AM Coughlin, D Miller, J and Kirk, S (2015) ‘The production of quick scoping reviews and rapid evidence assessments: a how to guide’, London: Department for Environment, Food and Rural Affairs https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/560521/Production_of_quick_scoping_reviews_and_rapid_evidence_assessments.pdf ↩

26. Office for National Statistics (2021) ‘Families and households in the UK: 2020’ (viewed on 14 March 2022) https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/families/bulletins/familiesandhouseholds/2020 ↩

27. Office for National Statistics (2021) ‘Families and households in the UK: 2020’ (viewed on 14 March 2022) https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/families/bulletins/familiesandhouseholds/2020 ↩

28. The Office for National Statistics defines a household as “one person living alone, or a group of people (not necessarily related) living at the same address who share cooking facilities and share a living room, sitting room and dining area. A household can consist of more than one family, or no families in the case of a group of unrelated people” (viewed on 25 March 2022) https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/families/bulletins/familiesandhouseholds/2012-11-01 ↩

29. This figure refers to all households of all types. ↩

30. Office for National Statistics (2021) ‘Families and households in the UK: 2020’ (viewed on 14 March 2022) https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/families/bulletins/familiesandhouseholds/2020 ↩

31. Office for National Statistics (2021) ‘Families and households in the UK: 2020’ (viewed on 14 March 2022) https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/families/bulletins/familiesandhouseholds/2020 ↩

32. Falkingham, J, Evandrou, M and Vlachantoni, A (2014) ‘Exploring the link between demographic change and poverty in the UK’, ESRC Centre for Population Change Working Paper 54, Southampton: University of Southampton (viewed on 04 May 2022) https://eprints.soton.ac.uk/368121/ ↩

33. Chzhen, Y and Bradshaw, J (2012) ‘Lone parents, poverty and policy in the European Union’, Journal of European Social Policy, 22 (5), 487-506 ↩

34. Eurostat (2021) ‘Household composition statistics – statistics explained’ (viewed on 14 March 2022) https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Household_composition_statistics ↩

35. Office for National Statistics (2021) ‘Families and households in the UK: 2020’ (viewed on 14 March 2022) https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/families/bulletins/familiesandhouseholds/2020 ↩

36. Eurostat (2021) ‘Household composition statistics – statistics explained’ (viewed on 14 March 2022) https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Household_composition_statistics ↩

37. Office for National Statistics (2021) ‘Population estimates by marital status and living arrangements, England and Wales: 2020’ (viewed on 14 March 2022) https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/bulletins/populationestimatesbymaritalstatusandlivingarrangements/2020 ↩

38. Office for National Statistics (2021) ‘Population estimates by marital status and living arrangements, England and Wales: 2020’ (viewed on 14 March 2022) https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/datasets/populationestimatesbymaritalstatusandlivingarrangements ↩

39. Habartová, P (2018) ‘Recent household trends in Europe: a cross-country analysis’, Demografie, 60 (2), 98-110 ↩

40. Office for National Statistics (2021) ‘Population estimates by marital status and living arrangements, England and Wales: 2020’ (viewed on 14 March 2022) https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/datasets/populationestimatesbymaritalstatusandlivingarrangements ↩

41. Falkingham, J, Evandrou, M and Vlachantoni, A (2014) ‘Exploring the link between demographic change and poverty in the UK’, ESRC Centre for Population Change Working Paper 54, Southampton: University of Southampton (viewed on 04 May 2022) https://eprints.soton.ac.uk/368121/ ↩

42. Single parent families are formed by a parent raising one or more children living in a household without a partner. Single parenthood typically results from separation, divorce or the death of a parent. Other factors include the absence of a parent for prolonged periods (e.g., due to migration), unintended pregnancy or the choice to raise a child alone. From: Jordan, V and Janta, B (2019) ‘Mechanisms supporting single parents across the European Union’, Luxembourg: Publications Office of the European Union (viewed on 6 April 2022) https://ec.europa.eu/social/main.jsp?catId=738&langId=en&pubId=8234&furtherPubs=yes ↩

43. A blended family is formed by a couple who have one or more children together, as well as (a) child(ren) from previous relationships. ↩

44. Picken, N and Janta, B (2019) ‘Leave policies and practice for non-traditional families’, Luxembourg: Publications Office of the European Union (viewed on 6 April 2022) https://ec.europa.eu/social/main.jsp?catId=738&langId=en&pubId=8239&furtherPubs=yes ↩

45. Carter, J (2012) ‘What is commitment? Women’s accounts of intimate attachment’, Families, Relationships and Societies, 1 (2), 137-153 ↩

46. Perelli-Harris, B and Bernardi, L (2015) ‘Exploring social norms around cohabitation: the life course, individualization, and culture’, Demographic Research, 33, 701-732 ↩

47. Office for National Statistics (2021) ‘Marriages in England and Wales’ (viewed on 14 March 2022) https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/marriagecohabitationandcivilpartnerships/datasets/marriagesinenglandandwales2013 ↩

48. British Social Attitudes 30 – personal relationships ↩

49. British Social Attitudes 30 – personal relationships ↩

50. Beaujouan, É and Bhrolcháin, MN (2011) ‘Cohabitation and marriage in Britain since the 1970s’, Population Trends, 145 (1), 35-59 ↩

51. Office for National Statistics (2021) ‘Marriages in England and Wales’ (viewed on 14 March 2022) https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/marriagecohabitationandcivilpartnerships/datasets/marriagesinenglandandwales2013 ↩

52. Beaujouan, É and Bhrolcháin, MN (2011) ‘Cohabitation and marriage in Britain since the 1970s’, Population Trends, 145 (1), 35-59 ↩

53. Hiekel, N Liefbroer, AC and Poortman, AR (2014) ‘Understanding diversity in the meaning of cohabitation across Europe’, European Journal of Population, 30 (4), 391-410. ↩

54. With the exception of Italians and Poles, who advised marriage before childbearing. ↩

55. Office for National Statistics (2010) ‘Do partnerships last? Comparing marriage and cohabitation using longitudinal census data’, Newport: Office for National Statistics (viewed on 14 March 2022) https://calls.ac.uk/wp-content/uploads/PopTrends04_tcm77-161305.pdf ↩

56. Office for National Statistics (2010) ‘Do partnerships last? Comparing marriage and cohabitation using longitudinal census data’, Newport: Office for National Statistics (viewed on 14 March 2022) https://calls.ac.uk/wp-content/uploads/PopTrends04_tcm77-161305.pdf ↩

57. Institute for Family Studies (2016) ‘Cohabitation, marriage, and union instability in Europe’ (viewed on 14 March 2022)https://ifstudies.org/blog/cohabitation-marriage-and-union-instability-in-europe ↩

58. Institute for Family Studies (2016) ‘Cohabitation, marriage, and union instability in Europe’ (viewed on 14 March 2022)https://ifstudies.org/blog/cohabitation-marriage-and-union-instability-in-europe ↩

59. Centre for Population Change (2015) ‘Cohabitation trends and patterns in the UK’, Southampton: Centre for Population Change (viewed on 14 March 2022) http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.728.8517&rep=rep1&type=pdf ↩

60. ) Centre for Population Change (2015) Cohabitation trends and patterns in the UK’, Southampton: Centre for Population Change (viewed on 14 March 2022) http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.728.8517&rep=rep1&type=pdf ↩

61. Relate (2017) It takes two: couples relationships in the UK’, Doncaster: Relate (viewed on 14 March 2022) [https://www.relate.org.uk/sites/default/files/the_way_we_are_now_-it_takes_two.pdf](https://www.relate.org.uk/sites/default/files/the_way_we_are_now-_it_takes_two.pdf) ↩

62. Office for National Statistics (2021) ‘Divorces in England and Wales: 2020’ (viewed on 14 March 2022) https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/divorce/bulletins/divorcesinenglandandwales/2020 ↩

63. Guzzo, KB (2014) ‘Trends in cohabitation outcomes: compositional changes and engagement among never‐married young adults’, Journal of Marriage and Family, 76 (4), 826-842 ↩

64. Guzzo, KB (2014) ‘Trends in cohabitation outcomes: compositional changes and engagement among never‐married young adults’, Journal of Marriage and Family, 76 (4), 826-842 ↩

65. Institute for Family Studies (2018) ‘Why is divorce declining in the UK?’ (viewed on 14 March 2022) https://ifstudies.org/blog/why-is-divorce-declining-in-the-uk ↩

66. Office for National Statistics (2021) ‘Births in England and Wales: summary tables’ (viewed on 14 March 2022) https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/livebirths/datasets/birthsummarytables ↩

67. Picken, N and Janta, B (2019) ‘Leave policies and practice for non-traditional families’, Luxembourg: Publications Office of the European Union (viewed on 6 April 2022) https://ec.europa.eu/social/main.jsp?catId=738&langId=en&pubId=8239&furtherPubs=yes ↩

68. Park, A, Bryson, C, Clery, E, Curtice, J and Phillips, M (eds) (2013) ‘British social attitudes: the 30th report’, London: National Centre for Social Research (viewed on 25 March 2022) https://www.bsa.natcen.ac.uk/media/1146/bsa30_personal_relationships_final.pdf ↩

69. Curtice, J, Hudson, N and Montagu, I (eds) (2020) ‘British social attitudes: the 37th report’, London: National Centre for Social Research (viewed on 25 March 2022) https://www.bsa.natcen.ac.uk/media/39410/bsa37_family-life.pdf ↩

70. ‘Living apart together’ refers to couples who self-identify as being in a relationships without living in the same household. ↩

71. Duncan, S and Phillips, M (2008) ‘New families? Tradition and change in partnering and relationships’, in Park, A, Curtice, J, Thomson, K, Phillips, M, Johnson, MC and Clery, E (eds) British Social Attitudes: the 24th Report, London: SAGE Publications ↩

72. Duncan, S, Phillips, M, Roseneil, S, Carter, J and Stoilova, M (2013) Living apart together: uncoupling intimacy and co-residence’, Sociology Teacher, 3 (1), 4-11 ↩

73. Duncan, S, Phillips, M, Roseneil, S, Carter, J and Stoilova, M (2013) Living apart together: uncoupling intimacy and co-residence’, Sociology Teacher, 3 (1), 4-11 ↩

74. Duncan, S, Phillips, M, Roseneil, S, Carter, J and Stoilova, M (2013) Living apart together: uncoupling intimacy and co-residence’, Sociology Teacher, 3 (1), 4-11 ↩

75. Duncan, S, Phillips, M, Roseneil, S, Carter, J and Stoilova, M (2013) Living apart together: uncoupling intimacy and co-residence’, Sociology Teacher, 3 (1), 4-11 ↩

76. Upton-Davis, K (2012) ‘Living apart together relationships (LAT): severing intimacy from obligation’, Gender Issues, 29 (1-4) 25-38 ↩

77. Reimondos, A, Evans, A and Gray, E (2011) ‘Living-apart-together (LAT) relationships in Australia’, Family Matters, 87, 43-55 ↩

78. Liefbroer, AC Poortman, AR and Seltzer, JA (2015) ‘Why do intimate partners live apart? Evidence on LAT relationships across Europe’, Demographic Research, 32, 251-286 ↩

79. Liefbroer, AC Poortman, AR and Seltzer, JA (2015) ‘Why do intimate partners live apart? Evidence on LAT relationships across Europe’, Demographic Research, 32, 251-286 ↩

80. Steinbach, A, Kuhnt, A-K and Knüll, M (2016) ‘The prevalence of single-parent families and stepfamilies in Europe: can the Hajnal line help us to describe regional patterns?’, History of the Family, 21, 4, 578-595 ↩

81. Steinbach, A, Kuhnt, A-K and Knüll, M (2016) ‘The prevalence of single-parent families and stepfamilies in Europe: can the Hajnal line help us to describe regional patterns?’, History of the Family, 21, 4, 578-595 ↩

82. Steinbach, A, Kuhnt, A-K and Knüll, M (2016) ‘The prevalence of single-parent families and stepfamilies in Europe: can the Hajnal line help us to describe regional patterns?’, History of the Family, 21, 4, 578-595 ↩

83. Halla, M, Lackner, M and Scharler, J (2016) ‘Does the welfare state destroy the family? Evidence from OECD member countries’, Scandinavian Journal of Economics, 118 (2), 292-323 ↩

84. Halla, M, Lackner, M and Scharler, J (2016) ‘Does the welfare state destroy the family? Evidence from OECD member countries’, Scandinavian Journal of Economics, 118 (2), 292-323 ↩

85. Halla, M, Lackner, M and Scharler, J (2016) ‘Does the welfare state destroy the family? Evidence from OECD member countries’, Scandinavian Journal of Economics, 118 (2), 292-323 ↩

86. Glaser, K, Stuchbury, R, Price, D, Di Gessa, G, Ribe, E and Tinker, A (2018) ‘Trends in the prevalence of grandparents living with grandchild(ren) in selected European countries and the United States’, European Journal of Ageing, 15 (3), 237-250 ↩

87. ) Glaser, K, Stuchbury, R, Price, D, Di Gessa, G, Ribe, E and Tinker, A (2018) ‘Trends in the prevalence of grandparents living with grandchild(ren) in selected European countries and the United States’, European Journal of Ageing, 15 (3), 237-250 ↩

88. Glaser, K, Stuchbury, R, Price, D, Di Gessa, G, Ribe, E and Tinker, A (2018) ‘Trends in the prevalence of grandparents living with grandchild(ren) in selected European countries and the United States’, European Journal of Ageing, 15 (3), 237-250 ↩

89. Cross-Barnet, C, Cherlin, A and Burton, L (2011) ‘Bound by children: intermittent cohabitation and living together apart,’ Family Relations, 60 (5), 633-647 ↩

90. Cross-Barnet, C, Cherlin, A and Burton, L (2011) ‘Bound by children: intermittent cohabitation and living together apart,’ Family Relations, 60 (5), 633-647 ↩

91. Cross-Barnet, C, Cherlin, A and Burton, L (2011) ‘Bound by children: intermittent cohabitation and living together apart,’ Family Relations, 60 (5), 633-647 ↩

92. Cross-Barnet, C, Cherlin, A and Burton, L (2011) ‘Bound by children: intermittent cohabitation and living together apart,’ Family Relations, 60 (5), 633-647 ↩

93. The system that preceded Universal Credit. ↩

94. Eurostat (2022) ‘At-risk-of-poverty rate by poverty threshold and household type – EU-SILC and ECHP surveys [ilc_li03]’, (viewed on 14 March 2022) https://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=ilc_li03&lang=en ↩

95. Antonelli, MA and De Bonis, V (2021) ‘Economic poverty: does the break-up of families matter?’ Social Sciences, 10 (6), 224 ↩

96. Falkingham, J, Evandrou, M and Vlachantoni, A (2014) ‘Exploring the link between demographic change and poverty in the UK’, ESRC Centre for Population Change Working Paper 54, Southampton: University of Southampton ↩

97. Iacovu, M (2017) ‘Household structure, income poverty and subjective hardship’, in Atkinson, AB, Guio, AC and Marlier, E (eds) (2017) ‘Monitoring social inclusion in Europe’, Luxembourg: Publications Office of the European Union ↩

98. Jordan, V and Janta, B (2019) ‘Mechanisms supporting single parents across the European Union’, Luxembourg: Publications Office of the European Union (viewed on 6 April 2022) https://op.europa.eu/en/publication-detail/-/publication/3ade5c22-b4d7-11e9-9d01-01aa75ed71a1/language-en ↩

99. Antonelli, MA and De Bonis, V (2021) ‘Economic poverty: does the break-up of families matter?’ Social Sciences, 10 (6), 224 ↩

100. Antonelli, MA and De Bonis, V (2021) ‘Economic poverty: does the break-up of families matter?’ Social Sciences, 10 (6), 224 ↩

101. This still includes the UK, as the analysis used 2017 data. ↩

102. Jordan, V and Janta, B (2019) ‘Mechanisms supporting single parents across the European Union’, Luxembourg: Publications Office of the European Union (viewed on 6 April 2022) https://op.europa.eu/en/publication-detail/-/publication/3ade5c22-b4d7-11e9-9d01-01aa75ed71a1/language-en; Chzhen, Y and Bradshaw, J (2012) ‘Lone parents, poverty and policy in the European Union’, Journal of European Commission Social Policy, 22 (5), 487-506 ↩

103. Iacovu, M (2017) Household structure, income poverty and subjective hardship, in Atkinson, AB, Guio, AC and Marlier, E (eds) (2017) ‘Monitoring social inclusion in Europe’, Luxembourg: Publications Office of the European Union ↩

104. Jordan, V and Janta, B (2019) ‘Mechanisms supporting single parents across the European Union’, Luxembourg: Publications Office of the European Union (viewed on 6 April 2022) https://op.europa.eu/en/publication-detail/-/publication/3ade5c22-b4d7-11e9-9d01-01aa75ed71a1/language-en ↩

105. Chzhen, Y and Bradshaw, J (2012) ‘Lone parents, poverty and policy in the European Union’, Journal of European Commission Social Policy, 22 (5), 487-506 ↩

106. Jordan, V and Janta, B (2019) ‘Mechanisms supporting single parents across the European Union’, Luxembourg: Publications Office of the European Union (viewed on 6 April 2022) https://op.europa.eu/en/publication-detail/-/publication/3ade5c22-b4d7-11e9-9d01-01aa75ed71a1/language-en ↩

107. Chzhen, Y and Bradshaw, J (2012) ‘Lone parents, poverty and policy in the European Union’, Journal of European Commission Social Policy, 22 (5), 487-506 ↩

108. Dermott, E and Pantazis, C (2014) ‘Poverty and Gender: Initial Findings PSE 2012’, Poverty and Social Exclusion ↩

109. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group ↩

110. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group ↩

111. Ponthieux, S (2017) ‘Intra‑household pooling and sharing of resources: a tentative “modified” equivalised income’, in Atkinson, AB, Guio, AC and Marlier, E (eds) Monitoring social inclusion in Europe, Luxembourg: Publications Office of the European Union ↩

112. Howard, M and Bennett, F (2021) ‘Distribution of money within the household and current social security issues for couples in the UK’, London: Women’s Budget Group ↩

113. Ponthieux, S (2017) ‘Intra‑household pooling and sharing of resources: a tentative “modified” equivalised income’, in Atkinson, AB, Guio, AC and Marlier, E (eds) Monitoring social inclusion in Europe, Luxembourg: Publications Office of the European Union ↩

114. Ponthieux, S (2017) ‘Intra‑household pooling and sharing of resources: a tentative “modified” equivalised income’, in Atkinson, AB, Guio, AC and Marlier, E (eds) Monitoring social inclusion in Europe, Luxembourg: Publications Office of the European Union ↩

115. Sonnenberg, SJ, Burgoyne, CB and Routh, DA (2011) ‘Income disparity and norms relating to intra-household financial organisation in the UK: a dimensional analysis’, Journal of Socio-Economics, 40 (5), 573-582 ↩

116. Ponthieux, S (2013) ‘Income pooling and equal sharing within the household – what can we learn from the 2010 EU-SILC module’, Eurostat methodologies and working papers, Luxembourg: Publications Office of the European Union ↩

117. Ponthieux, S (2013) ‘Income pooling and equal sharing within the household – what can we learn from the 2010 EU-SILC module’, Eurostat methodologies and working papers, Luxembourg: Publications Office of the European Union ↩

118. Barlow, A, Burgoyne, C and Smithson, J (2007) ‘The living together campaign – an investigation of its impact on legally aware cohabitants’, London: Ministry of Justice ↩

119. Picken, N and Janta, B (2019) ‘Leave policies and practice for non-traditional families’, Luxembourg: Publications Office of the European Union (viewed on 6 April 2022) https://ec.europa.eu/social/main.jsp?catId=738&langId=en&pubId=8239&furtherPubs=yes ↩

120. Ponthieux, S (2013) ‘Income pooling and equal sharing within the household – what can we learn from the 2010 EU-SILC module’, Eurostat methodologies and working papers, Luxembourg: Publications Office of the European Union ↩

121. Kelly, S (2007) ‘Ruling on cohabitation: a critical study of the cohabitation rule in UK social security law’, PhD thesis, University of Edinburgh (viewed on 25 March 2022) https://era.ed.ac.uk/handle/1842/29192 ↩

122. According to the British Social Attitudes Survey (Gender Roles edition), in the mid-1980s, close to half the public agreed that is “a man’s job is to earn money; a woman’s job is to look after the home and family”, while just 13% subscribed to this view in 2013. This decline is primarily a result of generational replacement, with new generations adhering less to traditional gender roles. ↩

123. Ponthieux, S (2017) ‘Intra‑household pooling and sharing of resources: a tentative “modified” equivalised income’, in Atkinson, AB, Guio, AC and Marlier, E (eds) Monitoring social inclusion in Europe, Luxembourg: Publications Office of the European Union ↩

124. Ponthieux, S (2017) ‘Intra‑household pooling and sharing of resources: a tentative “modified” equivalised income’, in Atkinson, AB, Guio, AC and Marlier, E (eds) Monitoring social inclusion in Europe, Luxembourg: Publications Office of the European Union ↩

125. Ponthieux, S (2017) ‘Intra‑household pooling and sharing of resources: a tentative “modified” equivalised income’, in Atkinson, AB, Guio, AC and Marlier, E (eds) Monitoring social inclusion in Europe, Luxembourg: Publications Office of the European Union ↩

126. Bennett, F and Sung, S (2013), ‘Gender implications of UK welfare reform and government equality duties: evidence from qualitative studies’, Oñati Socio-Legal Series, 3 (7), 1202-1221 ↩

127. Lechene, V, Pendakur, K and Wolf, A (2019) ‘OLS estimation of the intra-household distribution of consumption’, Working Paper W19/19, London: Institute for Fiscal Studies ↩

128. Bennett, F and Sung, S (2013), ‘Gender implications of UK welfare reform and government equality duties: evidence from qualitative studies’, Oñati Socio-Legal Series, 3 (7), 1202-1221 ↩

129. Bennett, F and Sung, S (2013), ‘Gender implications of UK welfare reform and government equality duties: evidence from qualitative studies’, Oñati Socio-Legal Series, 3 (7), 1202-1221 ↩

130. Bennett, F and Sung, S (2013), ‘Gender implications of UK welfare reform and government equality duties: evidence from qualitative studies’, Oñati Socio-Legal Series, 3 (7), 1202-1221 ↩

131. Bennett, F and Sung, S (2013), ‘Gender implications of UK welfare reform and government equality duties: evidence from qualitative studies’, Oñati Socio-Legal Series, 3 (7), 1202-1221 ↩

132. Marchal, S and Van Lancker, W (2019) ‘The measurement of targeting design in complex welfare states: a proposal and empirical applications’, Social Indicators Research, 143, 693-726 ↩

133. The unit that is used for the purposes of calculating benefit entitlement or payment. ↩

134. McLaughlin, E, Yeates, N and Kelly, G (2001) ‘Social security units of assessment: an international survey of the UK, Netherlands, Republic of Ireland and Australia and its implications for UK policy reform’, London: Trades Union Congress ↩

135. Eardley, T (2013) ‘Means-testing for social assistance: UK policy in an international perspective’, in Coyle, D, Lunt, N, McLaverty, P and Sorensen, J (eds) Welfare and policy: issues and agendas, London: Taylor and Francis, page 58-77 ↩

136. Eardley, T (2013) ‘Means-testing for social assistance: UK policy in an international perspective’, in Coyle, D, Lunt, N, McLaverty, P and Sorensen, J (eds) Welfare and policy: issues and agendas, London: Taylor and Francis, page 58 ↩

137. Eardley, T (2013) ‘Means-testing for social assistance: UK policy in an international perspective’, in Coyle, D, Lunt, N, McLaverty, P and Sorensen, J (eds) Welfare and policy: issues and agendas, London: Taylor and Francis, page 58 ↩

138. Eardley, T (2013) ‘Means-testing for social assistance: UK policy in an international perspective’, in Coyle, D, Lunt, N, McLaverty, P and Sorensen, J (eds) Welfare and policy: issues and agendas, London: Taylor and Francis, page 58 ↩

139. For comparative statistics on the proportion of the single parent households, see OECD (2011) ‘OECD Family Database’ (viewed on 28 March 2022) https://www.oecd.org/els/family/database.htm ↩

140. McLaughlin, E, Yeates, N and Kelly, G (2001) ‘Social security units of assessment: an international survey of the UK, Netherlands, Republic of Ireland and Australia and its implications for UK policy reform’, London: Trades Union Congress ↩

141. Department for Work and Pensions (2010) ‘Universal credit: welfare that works’ (viewed on 25 March 2022) https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/48897/universal-credit-full-document.pdf ↩

142. UK Government (2021) ‘Universal credit: further information for couples’ (viewed on 25 March 2022) https://www.gov.uk/government/publications/universal-credit-and-couples-an-introduction/universal-credit-further-information-for-couples ↩

143. UK Government (2013) ‘Advice for decision makers: staff guide’, Chapter E4, universal credit – Living Together as a Married Couple (viewed on 6 May 2022) https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/661551/adme4.pdf ↩

144. However, people living at the same dwelling do not necessarily form a household, for instance lodgers; UK Government (2013) ‘Advice for decision makers: staff guide’, Chapter E4, Universal credit – Living Together as a Married Couple (viewed on 6 May 2022) https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/661551/adme4.pdf ↩

145. UK Government (2013) ‘Advice for decision makers: staff guide’, Chapter E4, Universal credit – Living Together as a Married Couple (viewed on 6 May 2022) https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/661551/adme4.pdf ↩

146. UK Government (2013) ‘Advice for decision makers: staff guide’, Chapter E4, Universal credit – Living Together as a Married Couple (viewed on 6 May 2022) https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/661551/adme4.pdf ↩

147. UK Government (2013) ‘Advice for decision makers: staff guide’, Chapter E4, Universal credit – Living Together as a Married Couple (viewed on 6 May 2022) https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/661551/adme4.pdf ↩

148. UK Government (2013) ‘Advice for decision makers: staff guide’, Chapter E4, Universal credit – Living Together as a Married Couple (viewed on 6 May 2022) https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/661551/adme4.pdf ↩

149. UK Government (2013) ‘Advice for decision makers: staff guide’, Chapter E4, Universal credit – Living Together as a Married Couple (viewed on 6 May 2022) https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/661551/adme4.pdf ↩

150. UK Government (2013) ‘Advice for decision makers: staff guide’, Chapter E4, Universal credit – Living Together as a Married Couple (viewed on 6 May 2022) https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/661551/adme4.pdf ↩

151. UK Government (2013) ‘Advice for decision makers: staff guide’, Chapter E4, Universal credit – Living Together as a Married Couple (viewed on 6 May 2022) https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/661551/adme4.pdf ↩

152. Griffiths, R (2016) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, PhD thesis ↩

153. Child Tax Credit, Housing Benefit, Income Support, Income-based Job Seeker’s Allowance (JSA), Income-related Employment and Support Allowance (ESA), Working Tax Credit. ↩

154. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group ↩

155. Kelly, S (2007) ‘Ruling on cohabitation: a critical study of the cohabitation rule in UK social security law’, PhD thesis, University of Edinburgh (viewed on 25 March 2022) https://era.ed.ac.uk/handle/1842/29192 ↩

156. Koch, I (2015) ‘“The state has replaced the man”: women, family homes, and the benefit system on a council estate in England’, Focaal, 73, 84-96 ↩

157. None of the identified sources focused on (claimant or official) error in living together policies. ↩

158. Hirsch, D (2012) ‘Does the tax and benefit system create a couple penalty?’ York: Joseph Rowntree Foundation (viewed on 25 March 2022 www.jrf.org.uk/sites/default/files/jrf/migrated/files/benefits-tax-families-full.pdf ↩

159. Healey, O and Curtin, J (2019) ‘Relationship status and the welfare system in Aotearoa New Zealand’, report prepared for the Peter McKenzie Project ↩

160. Kelly, S (2007) ‘Ruling on cohabitation: a critical study of the cohabitation rule in UK social security law’, PhD thesis, University of Edinburgh (viewed on 25 March 2022) https://era.ed.ac.uk/handle/1842/29192 ↩

161. Healey, O and Curtin, J (2019) ‘Relationship status and the welfare system in Aotearoa New Zealand’, report prepared for the Peter McKenzie Project ↩

162. Adam, S and Brewer, M (2010) ‘Couples penalties and premiums in the UK tax and benefit system’ (viewed on 25 March 2022) https://ifs.org.uk/publications/4856 ↩

163. In the academic literature and the media, this is sometimes referred to as a ‘couple penalty’. ↩

164. Hirsch, D (2012) ‘Does the tax and benefit system create a couple penalty?’ York: Joseph Rowntree Foundation (viewed on 25 March 2022) www.jrf.org.uk/sites/default/files/jrf/migrated/files/benefits-tax-families-full.pdf ↩

165. Hirsch, D (2012) ‘Does the tax and benefit system create a couple penalty?’ York: Joseph Rowntree Foundation (viewed on 25 March 2022) www.jrf.org.uk/sites/default/files/jrf/migrated/files/benefits-tax-families-full.pdf ↩

166. Brewer, M, Browne, J and Wenchao, J (2011) ‘Universal credit: a preliminary analysis’, Briefing Note 116, London: Institute for Fiscal Studies ↩

167. House of Lords Economic Affairs Committee (2020) ‘Universal credit isn’t working: proposal for reform’ (viewed on 25 March 2022) https://publications.parliament.uk/pa/ld5801/ldselect/ldeconaf/105/105.pdf ↩

168. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561 ↩

169. Griffiths, R (2020) ‘For better or for worse: does the UK means-tested social security system encourage partnership dissolution?’ Journal of Poverty and Social Justice, 28 (1), 79-98 ↩

170. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’ ↩

171. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561 ↩

172. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561, page 549 ↩

173. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’ ↩

174. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’ ↩

175. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561 ↩

176. Griffiths, R (2020) ‘For better or for worse: does the UK means-tested social security system encourage partnership dissolution?’ Journal of Poverty and Social Justice, 28 (1), 79-98 ↩

177. Kelly, S (2007) ‘Ruling on cohabitation: a critical study of the cohabitation rule in UK social security law’, PhD thesis, University of Edinburgh (viewed on 25 March 2022) https://era.ed.ac.uk/handle/1842/29192 ↩

178. Griffiths, R (2020) ‘For better or for worse: does the UK means-tested social security system encourage partnership dissolution?’ Journal of Poverty and Social Justice, 28 (1), 79-98 ↩

179. Kelly, S (2007) ‘Ruling on cohabitation: a critical study of the cohabitation rule in UK social security law’, PhD thesis, University of Edinburgh (viewed on 25 March 2022) https://era.ed.ac.uk/handle/1842/29192 ↩

180. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561 ↩

181. Griffiths, R (2020) ‘For better or for worse: does the UK means-tested social security system encourage partnership dissolution?’ Journal of Poverty and Social Justice, 28 (1), 79-98 ↩

182. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’, page 129 ↩

183. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561 ↩

184. Kelly, S (2007) ‘Ruling on cohabitation: a critical study of the cohabitation rule in UK social security law’, PhD thesis, University of Edinburgh (viewed on 25 March 2022) https://era.ed.ac.uk/handle/1842/29192 ↩

185. Koch, I (2015) ‘“The state has replaced the man”: women, family homes, and the benefit system on a council estate in England’, Focaal, 73, 84-96 ↩

186. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561 ↩

187. Griffiths, R (2020) ‘For better or for worse: does the UK means-tested social security system encourage partnership dissolution?’ Journal of Poverty and Social Justice, 28 (1), 79-98 ↩

188. Kelly, S (2007) ‘Ruling on cohabitation: a critical study of the cohabitation rule in UK social security law’, PhD thesis, University of Edinburgh (viewed on 25 March 2022) https://era.ed.ac.uk/handle/1842/29192 ↩

189. Koch, I (2015) ‘“The state has replaced the man”: women, family homes, and the benefit system on a council estate in England’, Focaal, 73, 84-96 ↩

190. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’ ↩

191. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’ ↩

192. Bennett, F and Sung, S (2013), ‘Gender implications of UK welfare reform and government equality duties: evidence from qualitative studies’, Oñati Socio-Legal Series, 3 (7), 1202-1221 ↩

193. Bennett, F and Sung, S (2013), ‘Gender implications of UK welfare reform and government equality duties: evidence from qualitative studies’, Oñati Socio-Legal Series, 3 (7), 1202-1221 ↩

194. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561 ↩

195. Griffiths, R (2020) ‘For better or for worse: does the UK means-tested social security system encourage partnership dissolution?’ Journal of Poverty and Social Justice, 28 (1), 79-98 ↩

196. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561 ↩

197. Griffiths, R (2020) ‘For better or for worse: does the UK means-tested social security system encourage partnership dissolution?’ Journal of Poverty and Social Justice, 28 (1), 79-98 ↩

198. Griffiths, R (2020) ‘For better or for worse: does the UK means-tested social security system encourage partnership dissolution?’ Journal of Poverty and Social Justice, 28 (1), 79-98, page 93 ↩

199. Griffiths, R (2020) ‘For better or for worse: does the UK means-tested social security system encourage partnership dissolution?’ Journal of Poverty and Social Justice, 28 (1), 79-98, page 93 ↩

200. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561, page 555 ↩

201. Kelly, S (2007) ‘Ruling on cohabitation: a critical study of the cohabitation rule in UK social security law’, PhD thesis, University of Edinburgh (viewed on 25 March 2022) https://era.ed.ac.uk/handle/1842/29192 ↩

202. Koch, I (2015) ‘“The state has replaced the man”: women, family homes, and the benefit system on a council estate in England’, Focaal, 73, 84-96 ↩

203. Domański, H and Przybysz, D (2007) ‘Educational homogamy in 22 European countries’, European Societies, 9 (4), 495-526 ↩

204. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561, page 555 ↩

205. Kelly, S (2007) ‘Ruling on cohabitation: a critical study of the cohabitation rule in UK social security law’, PhD thesis, University of Edinburgh (viewed on 25 March 2022) https://era.ed.ac.uk/handle/1842/29192 ↩

206. Kelly, S (2007) ‘Ruling on cohabitation: a critical study of the cohabitation rule in UK social security law’, PhD thesis, University of Edinburgh (viewed on 25 March 2022) https://era.ed.ac.uk/handle/1842/29192 ↩

207. Kelly, S (2007) ‘Ruling on cohabitation: a critical study of the cohabitation rule in UK social security law’, PhD thesis, University of Edinburgh (viewed on 25 March 2022) https://era.ed.ac.uk/handle/1842/29192 ↩

208. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561 ↩

209. Kelly, S (2007) ‘Ruling on cohabitation: a critical study of the cohabitation rule in UK social security law’, PhD thesis, University of Edinburgh (viewed on 25 March 2022) https://era.ed.ac.uk/handle/1842/29192 ↩

210. Koch, I (2015) ‘“The state has replaced the man”: women, family homes, and the benefit system on a council estate in England’, Focaal, 73, 84-96 ↩

211. Howard, M and Bennett, F (2021) ‘Distribution of money within the household and current social security issues for couples in the UK’, London: Women’s Budget Group ↩

212. UK Government (2022) ‘Claim child benefit’ (viewed on 5 April 2022) https://www.gov.uk/child-benefit ↩

213. UK Government (2022) ‘Claim child benefit’ (viewed on 5 April 2022) https://www.gov.uk/child-benefit ↩

214. Howard, M and Bennett, F (2021) ‘Distribution of money within the household and current social security issues for couples in the UK’, London: Women’s Budget Group ↩

215. Howard, M and Bennett, F (2021) ‘Distribution of money within the household and current social security issues for couples in the UK’, London: Women’s Budget Group ↩

216. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’ ↩

217. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group ↩

218. Howard, M and Bennett, F (2021) ‘Distribution of money with in the household and current social security issues for couples in the UK’, London: Women’s Budget Group ↩

219. Griffiths, R (2020) ‘For better or for worse: does the UK means-tested social security system encourage partnership dissolution?’ Journal of Poverty and Social Justice, 28 (1), 79-98 ↩

220. Bennett, F and Sung, S (2013), ‘Gender implications of UK welfare reform and government equality duties: evidence from qualitative studies’, Oñati Socio-Legal Series, 3 (7), 1202-1221 ↩

221. Howard, M and Bennett, F (2021) ‘Distribution of money with in the household and current social security issues for couples in the UK’, London: Women’s Budget Group ↩

222. Griffiths, R (2020) ‘For better or for worse: does the UK means-tested social security system encourage partnership dissolution?’ Journal of Poverty and Social Justice, 28 (1), 79-98 ↩

223. Howard, M and Bennett, F (2021) ‘Distribution of money with in the household and current social security issues for couples in the UK’, London: Women’s Budget Group ↩

224. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561, page 553 ↩

225. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561 ↩

226. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’ ↩

227. Howard, M and Bennett, F (2021) ‘Distribution of money with in the household and current social security issues for couples in the UK’, London: Women’s Budget Group ↩

228. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group ↩

229. Bennett, F and Sung, S (2013), ‘Gender implications of UK welfare reform and government equality duties: evidence from qualitative studies’, Oñati Socio-Legal Series, 3 (7), 1202-1221 ↩

230. Bennett, F and Sung, S (2013), ‘Gender implications of UK welfare reform and government equality duties: evidence from qualitative studies’, Oñati Socio-Legal Series, 3 (7), 1202-1221 ↩

231. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’, page 14 ↩

232. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’ ↩

233. Griffiths, R (2020) ‘For better or for worse: does the UK means-tested social security system encourage partnership dissolution?’ Journal of Poverty and Social Justice, 28 (1), 79-98 ↩

234. Howard, M and Bennett, F (2021) ‘Distribution of money with in the household and current social security issues for couples in the UK’, London: Women’s Budget Group ↩

235. Office for National Statistics (2020) ‘Domestic abuse victim characteristics, England and Wales: year ending March 2020’ (viewed on 25 March 2022) https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/articles/domesticabusevictimcharacteristicsenglandandwales/yearendingmarch2020 ↩

236. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group ↩

237. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group page 9 ↩

238. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group ↩

239. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’ ↩

240. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group ↩

241. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’ ↩

242. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group ↩

243. Howard, M and Bennett, F (2021) ‘Distribution of money within the household and current social security issues for couples in the UK’, London: Women’s Budget Group, London: Women’s Budget Group ↩

244. Howard, M and Bennett, F (2021) Distribution of Money with in the Household and Current Social Security Issues for Couples in the UK. ↩

245. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’ ↩

246. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’ ↩

247. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’ ↩

248. Griffiths, R (2020) ‘For better or for worse: does the UK means-tested social security system encourage partnership dissolution?’ Journal of Poverty and Social Justice, 28 (1), 79-98 ↩

249. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group ↩

250. Koch, I (2015) ‘“The state has replaced the man”: women, family homes, and the benefit system on a council estate in England’, Focaal, 73, 84-96 ↩

251. Kelly, S (2007) ‘Ruling on cohabitation: a critical study of the cohabitation rule in UK social security law’, PhD thesis, University of Edinburgh (viewed on 25 March 2022) https://era.ed.ac.uk/handle/1842/29192 ↩

252. Koch, I (2015) ‘“The state has replaced the man”: women, family homes, and the benefit system on a council estate in England’, Focaal, 73, 84-96 ↩

253. Bennett, F and Sung, S (2013), ‘Gender implications of UK welfare reform and government equality duties: evidence from qualitative studies’, Oñati Socio-Legal Series, 3 (7), 1202-1221 ↩

254. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’, page 8 ↩

255. McLaughlin, E, Yeates, N and Kelly, G (2001) ‘Social security units of assessment: an international survey of the UK, Netherlands, Republic of Ireland and Australia and its implications for UK policy reform’, London: Trades Union Congress ↩

256. Millar, J (2004) ‘Squaring the circle? Means testing and individualisation in the UK and Australia’, Social Policy and Society, 3 (1), 67-74 ↩

257. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561 ↩

258. Griffiths, R (2016) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, PhD thesis ↩

259. McLaughlin, E, Yeates, N and Kelly, G (2001) ‘Social security units of assessment: an international survey of the UK, Netherlands, Republic of Ireland and Australia and its implications for UK policy reform’, London: Trades Union Congress ↩

260. Millar, J (2004) ‘Squaring the circle? Means testing and individualisation in the UK and Australia’, Social Policy and Society, 3 (1), 67-74 ↩

261. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561 ↩

262. Griffiths, R (2016) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, PhD thesis ↩

263. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561 ↩

264. Griffiths, R (2016) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, PhD thesis ↩

265. Millar, J (2004) ‘Squaring the circle? Means testing and individualisation in the UK and Australia’, Social Policy and Society, 3 (1), 67-74 ↩

266. Millar, J (2004) ‘Squaring the circle? Means testing and individualisation in the UK and Australia’, Social Policy and Society, 3 (1), 67-74, page 72 ↩

267. Prenzler T (2011) ‘Welfare fraud in Australia: dimensions and issues’, Trends and Issues in Crime and Criminal Justice 421, Canberra: Australian Institute of Criminology (viewed on 25 March 2022) https://www.aic.gov.au/publications/tandi/tandi421 ↩

268. Saunders, P (1999) ‘Social security in Australia and New Zealand: means‐tested or just mean?’ Social Policy and Administration, 33 (5), 493-515 ↩

269. Millar, J (2004) ‘Squaring the circle? Means testing and individualisation in the UK and Australia’, Social Policy and Society, 3 (1), 67-74 ↩

270. Whiteford, P and Angenent, A (2001) ‘The Australian system of social protection: an overview’, Canberra: Department of Family and Community Services ↩

271. Millar, J (2004) ‘Squaring the circle? Means testing and individualisation in the UK and Australia’, Social Policy and Society, 3 (1), 67-74 ↩

272. Saunders, P (1995) ‘Improving work incentives in a means-tested welfare system: the 1994 Australian social security reforms’, Fiscal Studies, 16 (2), 45-70 ↩

273. Millar, J (2004) ‘Squaring the circle? Means testing and individualisation in the UK and Australia’, Social Policy and Society, 3 (1), 67-74 ↩

274. Australian Government (2022) ‘Guides to social policy law: social security guide’, 4.2.2, Benefits income test and limits (viewed on 25 March 2022) https://guides.dss.gov.au/social-security-guide/4/2/2 ↩

275. Millar, J (2004) ‘Squaring the circle? Means testing and individualisation in the UK and Australia’, Social Policy and Society, 3 (1), 67-74 ↩

276. Certain types of income are discounted, for instance income from most government benefits and child support. For more information, see: Services Australia (2022) ‘Income – JobSeeker payment’ (viewed on 25 March 2022) https://www.servicesaustralia.gov.au/income?context=51411 ↩

277. Australian Government (2022) ‘Guides to social policy law: social security guide’, 4.2.2, Benefits income test and limits’(viewed on 25 March 2022) https://guides.dss.gov.au/social-security-guide/4/2/2 ↩

278. Services Australia (2022) ‘Income test for JobSeeker payment’ (viewed on 25 March 2022) https://www.servicesaustralia.gov.au/income-test-for-jobseeker-payment?context=51411 ↩

279. Assets might include property and home contents, financial investments, sole traders, partnerships, private trusts and private companies. See: Services Australia (2022) ‘Asset type – JobSeeker payment’ (viewed on 25 March 2022) https://www.servicesaustralia.gov.au/asset-types?context=51411 ↩

280. Prenzler T (2011) ‘Welfare fraud in Australia: dimensions and issues’, Trends and Issues in Crime and Criminal justice  421, Canberra: Australian Institute of Criminology (viewed on 25 March 2022) https://www.aic.gov.au/publications/tandi/tandi421 ↩

281. Services Australia (2022) ‘Income and asset tests for JobSeeker payment’ (viewed on 25 March 2022) https://www.servicesaustralia.gov.au/income-and-assets-tests-for-jobseeker-payment?context=51411 ↩

282. Services Australia (2022) ‘Income test for JobSeeker payment’ (viewed on 25 March 2022) https://www.servicesaustralia.gov.au/income-test-for-jobseeker-payment?context=51411 ↩

283. Millar, J (2004) ‘Squaring the circle? Means testing and individualisation in the UK and Australia’, Social Policy and Society, 3 (1), 67-74 ↩

284. Millar, J (2004) ‘Squaring the circle? Means testing and individualisation in the UK and Australia’, Social Policy and Society, 3 (1), 67-74 ↩

285. Saunders, P (1995) ‘Improving work incentives in a means-tested welfare system: the 1994 Australian social security reforms’, Fiscal Studies, 16 (2), 45-70 ↩

286. Saunders, P (1995) ‘Improving work incentives in a means-tested welfare system: the 1994 Australian social security reforms’, Fiscal Studies, 16 (2), 45-70 ↩

287. Harbridge, R and Bagley, P (2002) ‘Social protection and labour market outcomes in Australia’ in Sarfati, H and Bonoli, G (eds), Labour market and social protection reforms in international perspective, Abingdon: Routledge ↩

288. Warburton, M, Vuong, I and Evert, H (1999) ‘An evaluation of the working nation income test changes for unemployed people’, Canberra: Department of Family and Community Services ↩

289. Millar, J (2004) ‘Squaring the circle? Means testing and individualisation in the UK and Australia’, Social Policy and Society, 3 (1), 67-74 ↩

290. Prenzler T (2011) ‘Welfare fraud in Australia: dimensions and issues’, Trends and Issues in Crime and Criminal Justice  421, Canberra: Australian Institute of Criminology (viewed on 25 March 2022) https://www.aic.gov.au/publications/tandi/tandi421 ↩

291. Services Australia (2019) ‘Annual Report 2018-2019’ (viewed on 25 March 2022) https://www.transparency.gov.au/annual-reports/services-australia/reporting-year/2018-2019-62 ↩

292. Wilcock, S (2019) ‘(De-)criminalizing welfare? The rise and fall of social security fraud prosecutions in Australia’, British Journal of Criminology, 59 (6), 1498-1519 ↩

293. Wilcock, S (2019) ‘(De-)criminalizing welfare? The rise and fall of social security fraud prosecutions in Australia’, British Journal of Criminology, 59 (6), 1498-1519 ↩

294. Prenzler T (2011) ‘Welfare fraud in Australia: dimensions and issues’, Trends and Issues in Crime and Criminal Justice  421, Canberra: Australian Institute of Criminology (viewed on 25 March 2022) https://www.aic.gov.au/publications/tandi/tandi421 ↩

295. Australian Government (2022) ‘Guides to social policy law: social security guide’, 1.1.M.120, Member of a couple (viewed on 25 March 2022) https://guides.dss.gov.au/social-security-guide/1/1/m/120 ↩

296. Australian Government (2022) ‘Guides to social policy law: social security guide’, 1.1.M.120, Member of a couple (viewed on 25 March 2022) https://guides.dss.gov.au/social-security-guide/1/1/m/120 ↩

297. Australian Government (2022) ‘Guides to social policy law: social security guide’, 2.2.5.10, Determining a de facto relationship (viewed on 25 March 2022) https://guides.dss.gov.au/social-security-guide/2/2/5/10 ↩

298. Australian Government (2022) ‘Guides to social policy law: social security guide’, 2.2.5.10, Determining a de facto relationship (viewed on 25 March 2022) https://guides.dss.gov.au/social-security-guide/2/2/5/10 ↩

299. Services Australia (2022) ‘Your relationship status’ (viewed on 25 March 2022) https://www.servicesaustralia.gov.au/your-relationship-status?context=22151 ↩

300. Millar, J (2004) ‘Squaring the circle? Means testing and individualisation in the UK and Australia’, Social Policy and Society, 3 (1), 67-74 ↩

301. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561 ↩

302. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group ↩

303. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group ↩

304. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group ↩

305. UK Parliament (2019) ‘Welfare policy in Northern Ireland’ (viewed on 25 March 2022) https://publications.parliament.uk/pa/cm201719/cmselect/cmniaf/2100/210004.htm#:~:text=In%20Northern%20Ireland%2C%20a%20couple,claims%20making%20use%20of%20them. ↩

306. The consultation received 100 responses, 70 of which were from organisations and the remaining 30 of which were from individuals. ↩

307. Scottish Government (2022) ‘Universal credit: split payments’ (viewed on 25 March 2022) https://www.gov.scot/binaries/content/documents/govscot/publications/foi-eir-release/2018/05/foi-18-01007/documents/foi-18-01007-paper-shared-stakeholders-prior-15-december-meeting-pdf/foi-18-01007-paper-shared-stakeholders-prior-15-december-meeting-pdf/govscot%3Adocument/FOI-18-01007%2B-%2BPaper%2Bshared%2Bwith%2Bstakeholders%2Bprior%2Bto%2B15%2BDecember%2Bmeeting.pdf ↩

308. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group ↩

309. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’ ↩

310. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group ↩

311. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group ↩

312. Howard, M (2018) ‘Universal credit and financial abuse: making the links’, London: Women’s Budget Group ↩

313. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’ ↩

314. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’ ↩

315. Griffiths, R, Wood, M Bennett, F and Millar, J (2020) ‘Uncharted territory: universal credit, couples and money’ ↩

316. Bennett, F and Sung, S (2013), ‘Gender implications of UK welfare reform and government equality duties: evidence from qualitative studies’, Oñati Socio-Legal Series, 3 (7), 1202-1221 ↩

317. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561 ↩

318. Griffiths, R (2016) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, PhD thesis ↩

319. Griffiths, R (2017) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, Journal of Social Policy, 46 (3), 543-561 ↩

320. Griffiths, R (2016) ‘No love on the dole: the influence of the UK means-tested welfare system on partnering and family structure’, PhD thesis ↩

321. Ministry of Social Development (2013) ‘Tackling welfare fraud and welfare debt recovery: regulatory impact statement’, Wellington: Ministry of Social Development (viewed on 25 March 2022) https://www.treasury.govt.nz/publications/risa/regulatory-impact-statement-tackling-welfare-fraud-and-welfare-debt-recovery ↩

322. Eligibility for the following benefits is based on household income: New Zealand Superannuation, Student Allowance, Accommodation Supplement and Childcare and Out-of-School care subsidies, Temporary Additional Supplement, Working for Families tax credits, Winter Energy Payment; Fletcher, M (2018) ‘Individualising entitlements in New Zealand’s benefit and social assistance schemes: a report prepared for Superu’ (viewed on 25 March 2022) https://thehub.swa.govt.nz/assets/Uploads/NEW-Individualisation-of-benefits-FINAL.pdf ↩

323. Fletcher, M (2018) ‘Individualising entitlements in New Zealand’s benefit and social assistance schemes: a report prepared for Superu’ (viewed on 25 March 2022) https://thehub.swa.govt.nz/assets/Uploads/NEW-Individualisation-of-benefits-FINAL.pdf ↩

324. Healey, O and Curtin, J (2019) ‘Relationship status and the welfare system in Aotearoa New Zealand’, report prepared for the Peter McKenzie Project ↩

325. Fletcher, M (2018) ‘Individualising entitlements in New Zealand’s benefit and social assistance schemes: a report prepared for Superu’ (viewed on 25 March 2022) https://thehub.swa.govt.nz/assets/Uploads/NEW-Individualisation-of-benefits-FINAL.pdf ↩

326. St John, S, MacLennan, C, Anderson, H and Fountain, R (2014) ‘The complexities of “relationship” in the welfare system and the consequences for children’, Auckland: Child Poverty Action Group ↩

327. Fletcher, M (2018) ‘Individualising entitlements in New Zealand’s benefit and social assistance schemes: a report prepared for Superu’ (viewed on 25 March 2022) https://thehub.swa.govt.nz/assets/Uploads/NEW-Individualisation-of-benefits-FINAL.pdf ↩

328. Healey, O and Curtin, J (2019) ‘Relationship status and the welfare system in Aotearoa New Zealand’, report prepared for the Peter McKenzie Project 2019. ↩

329. Fletcher, M (2018) ‘Individualising entitlements in New Zealand’s benefit and social assistance schemes: a report prepared for Superu’ (viewed on 25 March 2022) https://thehub.swa.govt.nz/assets/Uploads/NEW-Individualisation-of-benefits-FINAL.pdf ↩

330. Healey, O and Curtin, J (2019) ‘Relationship status and the welfare system in Aotearoa New Zealand’, report prepared for the Peter McKenzie Project ↩

331. Healey, O and Curtin, J (2019) ‘Relationship status and the welfare system in Aotearoa New Zealand’, report prepared for the Peter McKenzie Project ↩

332. Healey, O and Curtin, J (2019) ‘Relationship status and the welfare system in Aotearoa New Zealand’, report prepared for the Peter McKenzie Project, page 11 ↩

333. Fletcher, M (2018) ‘Individualising entitlements in New Zealand’s benefit and social assistance schemes: a report prepared for Superu’ (viewed on 25 March 2022) https://thehub.swa.govt.nz/assets/Uploads/NEW-Individualisation-of-benefits-FINAL.pdf ↩

334. Healey, O and Curtin, J (2019) ‘Relationship status and the welfare system in Aotearoa New Zealand’, report prepared for the Peter McKenzie Project ↩

335. Fletcher, M (2018) ‘Individualising entitlements in New Zealand’s benefit and social assistance schemes: a report prepared for Superu’ (viewed on 25 March 2022) https://thehub.swa.govt.nz/assets/Uploads/NEW-Individualisation-of-benefits-FINAL.pdf ↩

336. Fletcher, M (2018) ‘Individualising entitlements in New Zealand’s benefit and social assistance schemes: a report prepared for Superu’ (viewed on 25 March 2022) https://thehub.swa.govt.nz/assets/Uploads/NEW-Individualisation-of-benefits-FINAL.pdf ↩

337. Fletcher, M (2018) ‘Individualising entitlements in New Zealand’s benefit and social assistance schemes: a report prepared for Superu’ (viewed on 25 March 2022) https://thehub.swa.govt.nz/assets/Uploads/NEW-Individualisation-of-benefits-FINAL.pdf ↩

338. Verification is not used to support the claim that two people are in a relationship. ↩

339. St John, S, MacLennan, C, Anderson, H and Fountain, R (2014) ‘The complexities of “relationship” in the welfare system and the consequences for children’, Auckland: Child Poverty Action Group ↩

340. Healey, O and Curtin, J (2019) ‘Relationship status and the welfare system in Aotearoa New Zealand’, report prepared for the Peter McKenzie Project ↩

341. St John, S, MacLennan, C, Anderson, H and Fountain, R (2014) ‘The complexities of “relationship” in the welfare system and the consequences for children’, Auckland: Child Poverty Action Group ↩

342. Marriott, L (2018) ‘Governing the poor: evidence from New Zealand’s tax and welfare systems’, Journal of New Zealand Studies, 27, 110-129 ↩

343. Marriott, L (2017) ‘The construction of crime: the presumption of blue-collar guilt and white-collar innocence’, Social Policy and Society, 16 (2), 237-251 ↩

344. St John, S, MacLennan, C, Anderson, H and Fountain, R (2014) ‘The complexities of “relationship” in the welfare system and the consequences for children’, Auckland: Child Poverty Action Group ↩

345. St John, S, MacLennan, C, Anderson, H and Fountain, R (2014) ‘The complexities of “relationship” in the welfare system and the consequences for children’, Auckland: Child Poverty Action Group ↩

346. Fletcher, M (2018) ‘Individualising entitlements in New Zealand’s benefit and social assistance schemes: a report prepared for Superu’ (viewed on 25 March 2022) https://thehub.swa.govt.nz/assets/Uploads/NEW-Individualisation-of-benefits-FINAL.pdf ↩

347. Healey, O and Curtin, J (2019) ‘Relationship status and the welfare system in Aotearoa New Zealand’, report prepared for the Peter McKenzie Project ↩

348. St John, S, MacLennan, C, Anderson, H and Fountain, R (2014) ‘The complexities of “relationship” in the welfare system and the consequences for children’, Auckland: Child Poverty Action Group ↩

349. St John, S, MacLennan, C, Anderson, H and Fountain, R (2014) ‘The complexities of “relationship” in the welfare system and the consequences for children’, Auckland: Child Poverty Action Group ↩

350. Marriott, L (2017) ‘The construction of crime: the presumption of blue-collar guilt and white-collar innocence’, Social Policy and Society, 16 (2), 237-251 ↩

351. Ministry of Social Development (2013) ‘Tackling welfare fraud and welfare debt recovery: regulatory impact statement’, Wellington: Ministry of Social Development’ (viewed on 25 March 2022) https://www.treasury.govt.nz/publications/risa/regulatory-impact-statement-tackling-welfare-fraud-and-welfare-debt-recovery ↩

352. Marriott, L (2017) ‘The construction of crime: the presumption of blue-collar guilt and white-collar innocence’, Social Policy and Society, 16 (2), 237-251 ↩

353. St John, S, MacLennan, C, Anderson, H and Fountain, R (2014) ‘The complexities of “relationship” in the welfare system and the consequences for children’, Auckland: Child Poverty Action Group ↩

354. Fletcher, M (2018) ‘Individualising entitlements in New Zealand’s benefit and social assistance schemes: a report prepared for Superu’ (viewed on 25 March 2022) https://thehub.swa.govt.nz/assets/Uploads/NEW-Individualisation-of-benefits-FINAL.pdf ↩

355. Fletcher, M (2018) ‘Individualising entitlements in New Zealand’s benefit and social assistance schemes: a report prepared for Superu’ (viewed on 25 March 2022) https://thehub.swa.govt.nz/assets/Uploads/NEW-Individualisation-of-benefits-FINAL.pdf ↩

356. Fletcher, M (2018) ‘Individualising entitlements in New Zealand’s benefit and social assistance schemes: a report prepared for Superu’ (viewed on 25 March 2022) https://thehub.swa.govt.nz/assets/Uploads/NEW-Individualisation-of-benefits-FINAL.pdf ↩

357. Fletcher, M (2018) ‘Individualising entitlements in New Zealand’s benefit and social assistance schemes: a report prepared for Superu’ (viewed on 25 March 2022) https://thehub.swa.govt.nz/assets/Uploads/NEW-Individualisation-of-benefits-FINAL.pdf ↩

358. Fletcher, M (2018) ‘Individualising entitlements in New Zealand’s benefit and social assistance schemes: a report prepared for Superu’ (viewed on 25 March 2022) https://thehub.swa.govt.nz/assets/Uploads/NEW-Individualisation-of-benefits-FINAL.pdf ↩

359. Healey, O and Curtin, J (2019) ‘Relationship status and the welfare system in Aotearoa New Zealand’, report prepared for the Peter McKenzie Project ↩

360. Fletcher, M (2018) ‘Individualising entitlements in New Zealand’s benefit and social assistance schemes: a report prepared for Superu’ (viewed on 25 March 2022) https://thehub.swa.govt.nz/assets/Uploads/NEW-Individualisation-of-benefits-FINAL.pdf ↩

361. Healey, O and Curtin, J (2019) ‘Relationship status and the welfare system in Aotearoa New Zealand’, report prepared for the Peter McKenzie Project ↩

362. Kelly, S (2007) ‘Ruling on cohabitation: a critical study of the cohabitation rule in UK social security law’, PhD thesis, University of Edinburgh (viewed on 25 March 2022) https://era.ed.ac.uk/handle/1842/29192 ↩

363. Kelly, S (2007) ‘Ruling on cohabitation: a critical study of the cohabitation rule in UK social security law’, PhD thesis, University of Edinburgh (viewed on 25 March 2022) https://era.ed.ac.uk/handle/1842/29192 ↩

364. Eardley, T (2013) ‘Means-testing for social assistance: UK policy in an international perspective’, in Coyle, D, Lunt, N, McLaverty, P and Sorensen, J (eds) Welfare and policy: issues and agendas, London: Taylor and Francis, page 58 ↩

365. The LTAMC policy aims to ensure that unmarried couples are treated in the same way as married couples. The LTAMC definition enables officials to recognise which relationships are ‘a couple’ for the purpose of administering universal credit and other benefits. The term LTAMC is not defined in legislation. Decision makers must determine on the basis of available evidence whether the entire relationship of two people who are not married to each other is comparable to that of a couple who are married. The LTAMC definition is complex because it considers several different dimensions, not all of which have to be present. The first consideration in deciding whether two people are LTAMC is whether they are members of the same household. The household is not defined in legislation. It is a domestic environment containing the essentials of home life. To be members of the same household, two people should share a dwelling and live there regularly. ↩

366. Due to the open nature of Google Scholar, it is necessary to place some restrictions on the number of results to be screened. Pilot searches indicate that limiting the screening exercise to the first 50 results should generate a sufficient number of relevant results, although we will assess this parameter as we conduct the search and update or adapt the approach as needed. ↩

367. NatCen (####) British Social Attitudes 38 (viewed on 25 March 2022) https://www.bsa.natcen.ac.uk/ ↩

368. ISSP (2012) ‘Family and changing gender roles IV’ (viewed on 25 March 2022) https://www.gesis.org/en/issp/modules/issp-modules-by-topic/family-and-changing-gender-roles/2012 ↩

369. Eurostat (2022) ‘Distribution of households by household type from 2003 onwards – EU-SILC survey [ilc_lvph02]’ (viewed on 25 March 2022) https://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=ilc_lvph02&lang=en ↩

370. Piloting the search terms suggested that some relevant sources for the UK may be missed from Google Scholar if we review only the first 30 results when we include the UK along with other countries. ↩