Timing it right or timing it wrong: how should income-tested benefits deal with changes in circumstances?

Winner – 2019 Best Paper Prize of the Foundation for International Studies on Social Security (FISS) sponsored by the Journal of Poverty and Social Justice

View author details View Less
  • 1 University of Bath, , UK
  • | 2 Australian National University, , Australia
Full Access
Get eTOC alerts
Rights and permissions Cite this article

This article examines the challenges in designing income-tested benefits for people of working age. This is particularly difficult in the context of changing patterns of work and volatility in earnings and income. Matching benefits to needs requires timely assessment and payment. We compare the treatment of timing issues in the working-age welfare systems of the United Kingdom and Australia. The article discusses how these different but similar systems deal with the timing of income receipt and benefit adjustment, problems of overpayment and debt, and draws out some lessons for the design of income-tested provisions.

Abstract

This article examines the challenges in designing income-tested benefits for people of working age. This is particularly difficult in the context of changing patterns of work and volatility in earnings and income. Matching benefits to needs requires timely assessment and payment. We compare the treatment of timing issues in the working-age welfare systems of the United Kingdom and Australia. The article discusses how these different but similar systems deal with the timing of income receipt and benefit adjustment, problems of overpayment and debt, and draws out some lessons for the design of income-tested provisions.

Introduction

Social security programmes respond to changes in circumstances, particularly falls in income, whether these are predictable due, for example, to retirement or taking time off paid work on the birth of a child, or for more unpredictable risks such as unemployment, short-term sickness, disability or caring. Social support becomes payable because of loss of earnings. Correspondingly, support ceases when individuals leave unemployment for sufficiently well-paid work, or their health improves and they can return to work. Social security payments thus stabilise household incomes, both in macroeconomic terms when there are recessions, or more narrowly when individuals or families experience personal, but common, risks (Barr, 1987).

One of the central design challenges of any income-tested welfare system is how to assess needs against income in a timely way, in order to determine how much benefit should be paid. This involves a series of decisions, including ‘what is taken into account as income, whose income in a family unit is counted, over what time period income is measured, how long awards should last, and how responsive the system should be to changes in income and circumstances during the period of the award’ (Whiteford et al, 2003: 1). These design issues have grown in policy importance in recent years. Labour market developments have made work more unstable and wages more variable for many people. Part-time and other non-standard forms of employment have increased in many developed economies, including more self-employment and multiple job holding (OECD, 2019a). There have also been increases in the most ‘precarious’ forms of employment, including temporary jobs, casual work, contract/agency employment, zero-hours contracts (Rubery et al, 2018), including in the UK (ONS, 2018) and Australia (Campbell and Burgess, 2018). There are also growing concerns about job losses as a result of the possible influence of automation (McKinnon, 2019).

Related to these changing patterns of employment are changes in the distribution of earnings, often associated with employment polarisation – the loss of middle-skilled jobs and the growth in both low-skilled and high-skilled jobs. There is growing evidence to suggest that earnings and income have become more unstable over relatively short periods (Hills et al, 2006; Morduch and Schneider, 2017; Campbell and Burgess, 2018; Hardy and Lane, 2018; Tomlinson, 2018). This leads to income instability, ‘changes in income that are unpredictable or unintentional, and that do not lead to improved economic circumstances’ (Romich and Hill, 2017: 1). This instability also creates additional demands on systems of social protection, specifically in respect of ensuring income adequacy for people in work, but with low and/or variable earnings.

There are thus growing concerns about whether existing systems of social protection are suited to these changing patterns of work (Clegg, 2013; Greve, 2019). Three main problems have been identified. First there is the coverage of social protection systems and whether these can be inclusive – and provide adequate protection – for people in a range of employment circumstances (IMF, 2018; World Bank, 2019). Second there are concerns about future funding and financing of systems, as OECD notes that ‘more volatile career patterns or a growing diversity of employment forms pose specific challenges for social protection provisions that link support entitlements or financing burdens to past or present employment’ (2019b: 295).

Third – and most relevant for our interest here – is the question of how to deal with people who have unstable or irregular employment, moving from job to job, between work and unemployment, or working in more than one job. Reforms to social security to address this employment volatility include measures to encourage labour force participation through stronger activation requirements and measures to encourage part-time work and job progression. Many countries have in-work benefits or tax credits intended to improve work incentives and to reduce the risk of in-work poverty (Millar, 2008; Clasen, 2019). These provisions put benefits alongside earnings, creating an ‘income package’ made up of different components and combinations. When income-tested provisions are in the mix, as they often are, this can create complex design and delivery problems.

The OECD has suggested that: ‘In order to ease access barriers to social protection, policy makers should consider: … making means tests more responsive to people’s needs by shortening the reference periods for needs assessments and by putting appropriate weight on recent or current incomes of all family members’ (2019a: 325).

Responsiveness, however, can be particularly challenging when changes in income and circumstances are frequent and unpredictable. There is a risk that payments get out of step with circumstances resulting in underpayments or overpayments, and hence debts to be repaid, which can lead to individual hardship and political controversy for governments. The costs of getting this wrong are potentially high, not just financially but also in relation to the perceived legitimacy of the system.

This article considers the cases of the UK and Australia, two countries marked by a high level of reliance on income testing (OECD, 2016). The UK’s experiences are of interest as this is a country that is currently introducing radical benefit reform for working-age households through the introduction of Universal Credit. Australia’s experiences are relevant, as aspects of its income testing are similar to Universal Credit, but have been in place for payments to the unemployed since the 1990s and for payments to lone parents and people with disabilities since the 1960s. We focus on rules for the income test and not the full means test, which may include taking account of savings and assets. This is because assets are a stock, whereas income is the more relevant flow, and because changes in income and circumstances are more common in practice than adjustments for changes in capital. We also focus on benefits for working people, those who are – or are seeking to – combine wages and benefits. We examine how the two countries have defined the relevant time periods for assessment and payment of income-tested benefits and how they deal with overpayments. The article has three main sections. The first and second discuss policy developments in the UK and Australia respectively. The final section draws out some lessons of these case studies for the design of income tests in social security provisions.

The UK: Tax Credits and Universal Credit

In the UK, the two main income-tested provisions for working people are the Working Tax Credit and Child Tax Credit, introduced in 2003. From 2013 there has been a staged introduction of Universal Credit, replacing these two tax credits and also four income-tested benefits (Housing Benefit, Income Support, and the means-tested versions of Jobseeker’s Allowance and of Employment and Support Allowance). The implementation of Universal Credit will not be complete until 2023/4, so both systems currently operate alongside each other. Which one a claimant can access depends partly on where they live, as some areas of the country are ahead of others in Universal Credit rollout, and new claimants in those areas must claim Universal Credit. There are also two other routes from the so-called ‘legacy benefits’ to Universal Credit. ‘Natural migration’ refers to people having a change of circumstances ending their legacy benefits and making a new claim to Universal Credit. ‘Managed migration’ starts in 2019 and refers to the process by which people on legacy benefits will have their claims transferred to Universal Credit (SSAC, 2018).

Here we focus on timing issues in the provisions for people in work: tax credits (annual assessment and four-weekly payment) and Universal Credit (monthly assessment and payment).1 Both systems have significant reach into the working-age population. About 4 million households (mainly families with children) receive tax credits (in 2017/8) and about 2.1 million households (mainly single people and couples without children) receive Universal Credit (in August 2019) (HMRC, 2019; DWP, 2019). Around 6.5 to 7 million households will be receiving Universal Credit when it is fully rolled out by 2023/4 (Kennedy and Keen, 2018).

Tax Credits

The Working Tax Credit is payable to working adults with or without children, working at least 16 hours per week for those in families with children and workers with a disability, or otherwise 30 hours per week. It includes a payment for childcare, meeting a proportion of costs up to a maximum. Child Tax Credit is payable to families with children, in and out of work. Both are administered by HM Revenue and Customs (HMRC).

The income test for tax credits starts with a retrospective assessment, in which the amount of the award is based on the previous tax year’s gross income. The period of the award is 12 months, with reconciliation at the end of 12 months between the amounts that should have been paid during the year, given any changes in income and circumstances, and what was actually paid. The default payment frequency is four-weekly, but recipients can opt for weekly payments. Some changes in circumstances are required to be reported within one month, for example a couple separating or starting to live together, childcare costs changing, being abroad for more than eight weeks, working hours reducing to fewer than 16 per week, or changing from above or below 30 hours, children leaving home or dying. Other changes in income and circumstances can either be reported immediately or reported at year-end for the annual reconciliation.

Over- and underpayments by design

The design of tax credits follows a ‘pay now, establish entitlement later’ approach (Revenue Benefits, 2019). This inevitably means that there are both underpayments and overpayments in the system. The initial assessment is provisional, and only finalised when the actual income for the year is known, at year-end. Most over- and underpayments are because changes in income have not been reported until the end of year reconciliation (HMRC, 2018). Other reasons are that claimants fail to report changes of circumstances until the year-end, or because changes are not reported in time to reduce payments in that year, or because of office error.

From 2003, there were substantial and very visible problems with overpayments, or more precisely visible problems in terms of the impact on claimants of having to repay overpayments. The number of overpayments, and their costs, were much higher than predicted. In the first year (2003/04), about one-third of all tax credits awards paid – nearly 1.9 million awards – were overpaid, at a cost of nearly £2 billion (House of Commons Treasury Select Committee, 2006). Media coverage reported confusion, hardship and debt as the government sought to recover overpayments. Research showed the importance of tax credits to family income, but also highlighted the negative impact of late and incorrect payments, payments that varied inexplicably, reductions in awards for overpayments, and the lack of detailed information about awards (Millar, 2008; 2011). There were critical reports from the House of Commons Treasury Select Committee (2006), the Institute for Fiscal Studies (Chote et al, 2006), the National Audit Office (2007) and the Parliamentary Ombudsman (2007).

Because of these concerns, and to reduce overpayments, the government substantially increased the disregard for income increases. In the original design the annual disregard for income rises was set at £2,500. From April 2006, this increased to £25,000, in order to recognise the volume and level of the changes. As Baroness Hollis (the Labour government minister responsible) explained:

It is fair to say that when we introduced the Tax Credits Bill we did not predict that 50 per cent of lone parents would undergo more than a dozen changes in circumstance a year. Those include changes in childcare arrangements virtually every school holiday, changes in hours worked and sometimes a change of partner. […] the Government [also] underestimated the occasions on which the female in a couple household went into work and produced a major increase in family income. (Lords Hansard, 23 October 2006: Column 1060)

It is intriguing to imagine how tax credits might have evolved subsequently, but time and political priorities moved rapidly and changed radically, not least with the Coalition (Conservative and Liberal Democrat) government in power after 2010. The introduction of Universal Credit moved to the top of the social security policy agenda, as we discuss later. More immediately, the government was seeking ways to reduce public expenditure. The generous disregard for income rises did not survive the new climate of austerity, and was among the elements targeted in the ‘emergency’ Budget in 2010. The disregard was reduced to £10,000 from April 2011, to £5,000 from April 2013, and back to the original amount of £2,500 from April 2016.2

This of course meant that overpayments started to rise again. By 2016–17, the profile for overpayments was almost the same as it had been in 2003/4, with one third of awards – around 1.6 million – overpaid, at a cost of around £1.7 billion (HMRC, 2018). But reducing these overpayments is apparently not now a priority for government, as noted by the House of Commons Committee of Public Accounts (2018: 9): ‘Tax Credits are being replaced […] and therefore the appetite within government to change existing systems and processes is very low.’ Attention now is all on Universal Credit, which has a very different approach to defining relevant time periods and to addressing changes in circumstances.

Universal Credit

As noted earlier, Universal Credit combines four existing benefits and two tax credits for working-aged people into one single system, paid in and out of work. It is based on a monthly assessment of income and needs, and then paid monthly in arrears. Because Universal Credit is paid in and out of work, eligibility continues as people move into work, or between jobs and changing hours of work. The system is highly IT-based by design, which requires initial claims to be made online for most people, and then online reporting of changes in circumstances and compliance with job search and other work requirements. Work requirements are central to the scheme, and are applied to most claimants, including both partners in a couple, with some variations and exemptions for caring responsibilities and disability, and there may be requirements for some working people to seek to increase their hours of work.

The calendar month is the key time period (Millar and Bennet, 2017; Tucker, 2018). Universal Credit is assessed and paid in arrears, on a monthly basis and in a single payment. The assessment period starts on the date that the claim is made and is set at one calendar month. The first payment follows seven days after the end of that first assessment period. This means that there is usually (if all goes smoothly) five weeks from the claim date to the first payment. Then payments are made on the same date each month. For most employees, earnings information is collected from employers using the ‘real-time information’ system. Self-employed people, and those whose earnings are not automatically uploaded to HMRC, must report earnings themselves on a monthly basis. Other changes in personal or family circumstances must be reported by the claimants and are applied on a ‘whole-month’ basis, which means that whatever circumstances apply on the last day of the assessment period are applied to the whole of the previous month, regardless of when they actually happened. The assessment also takes into account two main areas of expenditure, housing costs and childcare costs (for those eligible) and these are included in the monthly assessment period.

Thus, the calendar month is in effect a personalised and self-contained entity, in which income, circumstances and outgoings are calculated within that period, and payment follows. In theory, there should be no over or under payments, as the award is adjusted each month if earnings, family circumstances and relevant outgoings (for example, childcare costs or housing costs) change. This is intended to be a fully responsive system, making payments – adjusted as necessary – on an ongoing basis. But in practice monthly assessment and monthly payments have caused some significant problems in administration, and for claimants.

The monthly assessment period is not always lined up with wages, especially for lower-paid people who are the least likely to be paid monthly (Tucker and Norris, 2018). Weekly pay can translate into four or five pay packets, depending on the month. Not all those who are paid monthly are paid on the same day each month. Earnings can be highly irregular for self-employed people, and those on variable hours contracts. In two-earner couples, both partners may be paid on a different days. Any lump sums (back pay, holiday pay, tax rebates) are treated as income for the assessment period in which the payment is made, even if it relates to a previous period. Employers supply the information on earnings, which reduces the reporting requirements on claimants and should in theory mean that the Universal Credit payment can immediately and accurately be adjusted to reflect changes in earnings. But errors in the earnings information from employers are hard for claimants to identify, and can only be resolved between DWP and HMRC (Todd, 2018).

There are other complications. As noted earlier, self-employed people – who may often have irregular income – must report earnings themselves. For the first 12 months these reported earnings are used to calculate the Universal Credit payment. After 12 months the ‘minimum income floor’ applies. If someone earns below this they are treated as earning at the floor (which is set as 35 hours at the National Minimum Wage). If someone earns more than this then actual earnings are taken into account. The ‘surplus earnings’ rule applies to both self-employed people and employees and is intended to prevent claimants from manipulating their earnings to maximise their claims. These rules apply when someone reclaims within six months of a previous claim ending, with their ‘surplus’ from their previous claim (after a disregard) taken into account to reduce the current award. Charlesworth (2018) notes that these rules ‘require historical earnings information which will be tricky for advisors to obtain and explain, let alone claimants to understand’.

For all claimants, any other changes in income or circumstances must be reported via the online journal, which some people (about a fifth of claimants in the DWP (2018) survey) find difficult, especially those without internet access. Overpayments have not disappeared under Universal Credit. In 2019 of expenditure the ‘estimated overpayments of Universal Credit are the highest of currently measured benefits at 8.6% – the highest rate since 2002–03 when overpayments of Tax Credits (administered by HM Revenue & Customs) were 9.7%’ (National Audit Office, 2019: 10). The most common reason was ‘untimely and inaccurate’ information on income and earnings.

As noted earlier, in addition, the whole month assessment period might be good or bad for claimants, depending on when in the month a change happens (Millar and Bennet, 2017). But it also adds to the uncertainty, and can lead to ‘arbitrary financial shortfalls, purely because of the date when these events occurred’ (Tucker and Norris, 2018: 3).

There are also problems that follow from the monthly payment. These include well-documented financial problems caused by the five-week waiting period to the first payment and the challenges for some claimants in adapting to budgeting over a month (Tucker and Norris, 2018; JItendra et al, 2019). The way in which childcare costs are covered can also be a source of budgeting problems. Claimants are required to pay for childcare upfront and submit receipts after the childcare has been provided. Finding upfront costs can be challenging. Here again the whole month approach can be unhelpful: ‘A parent might pay upfront for a month’s worth of childcare in the middle of their UC assessment period. They would only receive money back at the end of that assessment period for care that had actually been provided. The remainder would be reimbursed in the following month’ (House of Commons Work and Pensions Select Committee, 2018: 14). There are also possible deductions from Universal Credit for reasons that relate, not to current, but to past circumstances. The Welfare Reform Act 2012 allows reciprocal sharing of HMRC and DWP information and this provides a mechanism for identifying tax credit overpayments for recovery. Deductions from Universal Credit include deductions for overpayments, advance payments, hardship payments, tax credit and other benefit overpayments, and for certain other debts including rent and utilities (Millar and Whiteford, 2017; Kennedy et al, 2019).

Bad timings?

The annual assessment approach of tax credits was criticised as being too long by the House of Commons Treasury Select Committee (2005): ‘the tax credits regime, which is designed to deliver the correct amount of State assistance over the year as a whole, rather than over any shorter period, could be aligned more closely to the financial needs of such families’. But the monthly assessment in Universal Credit is arguably too short and the whole month approach too rigid, yet at the same time the system is too responsive to what might be short-term changes. And the monthly payment is arguably too long for low-income budgeting. The fact that Universal Credit assessments are made every month means that payments can change from month to month, including in ways that are not necessarily predictable to the claimant, also making it difficult to budget. Overall, there is clearly a risk that, for a significant number of claimants, the relationship between needs and payments will be adrift and opaque.

These issues could have been foreseen. We quoted Baroness Hollis earlier, reflecting on the way the government had underestimated the sheer volume of changes of circumstances in people’s lives. But Baroness Hollis also warned of the risks inherent in being highly responsive to such changes:

[If] you seek to track every change and every three to four weeks change the credit for half the population claiming tax credits, even if the computer could handle it, I doubt very much whether the lone parent could. Such adjustments would be made six weeks in arrears and there would be no way in which that parent would be able to construct a family budget with such unreliable and non-robust flows of income, especially as some of the changes in circumstance cancel each other out. (Lords Hansard, 23 Oct 2006: Column 1060)

This now reads like a prescient comment on the challenges from design to delivery of Universal Credit.

Income-testing in the Australian system

Income-testing in Australia has differed significantly from practices in the UK – until the introduction of Universal Credit, at least. Over time, the Australian system evolved from a tightly income-tested social assistance system to one in which the income tests are similar to those for Universal Credit, but differ for people who fall into specific eligibility categories.3

The shape of the income tests is broadly similar in both countries. In Australia, there is a free area, below which payments are not reduced, broadly similar to the work allowance in Universal Credit, but applying to all income and available for all claimants. Above the free areas, the withdrawal rates are between 40 per cent and 60 per cent. (In the UK the work allowance applies only to people with children and those with limited work capability and the withdrawal rate for Universal Credit is 63 per cent.) Income support payment periods are generally fortnightly, and the regular assessment period is also 14 days, with payments being made either one or two days after the end of the assessment period. Payments on a fortnightly cycle align with most wage and salary earners in Australia, who also have fortnightly pay – although not necessarily on the same payment day.

If claimants experience a change in circumstances that affects their level of entitlement, they have 14 days from the date of the event to advise Centrelink, the benefit delivery agency. There are a number of mechanisms to adjust the impact of the income testing of earnings, when earnings are variable. Part of unused free areas accumulate automatically over time. This ‘Income Bank’ allows unemployed people to earn more before their benefit payment reduces when their income goes over $48 per fortnight. In fortnights with higher earnings, available Working Credits offset excess earnings until the Working Credit balance is zero, and then the income support payment starts to reduce. Thus, the Working Credit takes account of fluctuations in earnings by smoothing out the impact of these over time, and also reduces disparities caused by any differences between the time income is earned and received.

In March 2019, around 17 per cent of working-aged income support recipients had earnings as well as their income support payments (Department of Social Services, 2019). This varied from 39 per cent for recipients of the Youth Allowance (Student) to 29 per cent for single parents receiving Parenting Payment, 20 per cent for those on Newstart Allowance and 18 per cent for claimants of Youth Allowance (Other). In March 2019, there were 500,000 people both working and receiving social security payments, which was equivalent to around 12 per cent of all part-time workers in Australia. There are also many low- and middle-income families with children simultaneously receiving family payments and working.

Family Tax Benefits, overpayments and debts

Over time, Australia has developed an integrated system of cash assistance for working and non-working families with children. A modest programme (Family Income Supplement (FIS), based on the UK scheme of the same name) was introduced in 1983, providing assistance for low-income working families, at the same rates of payment per child as for income support recipients. From 1993, there was an integrated payment for low- to middle-income families with children, with the money usually paid directly to primary carers, typically mothers. Between 1983 and the late 1990s, coverage expanded from 1 per cent to nearly 14 per cent of children (Stewart and Whiteford, 2018).

In July 2000, the government introduced major changes to the tax system, accompanied by changes to family assistance, simplifying and increasing payments, and reducing income-test withdrawal rates. There was a wider range of payment choices. Families not receiving adult income support could choose to receive direct payments into bank accounts each fortnight, or choose to receive payments through the tax system, either as a lump sum or periodically by employers as a departure from their otherwise scheduled ‘Pay-As-You-Go’ (PAYG) tax instalment deductions (TIDs). Families not receiving adult income support were able to change their payment method at any time during the year. When introduced in 1983, entitlement to FIS was based on average joint parental ordinary income before tax in the four weeks preceding application. Entitlements were reviewed every six months. Recipients whose income fell could apply for reassessment at any time (and were required to notify the Department of Social Security if their four-weekly average income exceeded 125 per cent of the income on which the original assessment was made).

Under the system from 2000, families who chose to receive Family Tax Benefit (FTB) as a regular payment were asked to estimate their ‘adjusted’ taxable income for the next financial year in advance. There was an end-of-financial-year reconciliation for all families when payments received were assessed against what ought to have been paid. If actual income was more than had been estimated, there was an overpayment and a debt.

This end-of-year reconciliation reflected the primary objectives that families receive assistance when they need it and that families could choose the means of delivery, including through the tax system. Given that Australia’s income tax system involves submitting annual returns, used to reconcile under or overpayments of tax, this meant an extension of reconciliation to family assistance. Reconciliation was required to ensure that families with the same annual incomes would receive the same entitlement, irrespective of the delivery mechanism.

Controversy about the new system started towards the end of the 2000–2001 financial year, when the time first approached for reconciling entitlements. In 1999–2000, there had been about 51,800 families with debts at the end of the year. In contrast, after the introduction of the new system, an estimated 670,000 families were overpaid – more than 12 times as many as the previous year (Whiteford et al, 2003).4 In July 2001, just before an important by-election, the government announced that it would waive the first $1,000 of all overpayments.

Why did the level of overpayments increase so significantly? The main reason was that under the previous system entitlements had been based on family income during the previous tax year, whereas in the new system it was based on an estimate of future income.

The waiver did not resolve all problems. There was a rising number of complaints, leading to a report by the Commonwealth Ombudsman (2003). This identified the main areas of concern: that the system seemed inherently to result in a large number of debts; that many debts were large; that debts affected many lower-income families; that debts might be unavoidable, even when families fully complied with all requirements; and that debts seemed to have an unfair retrospective effect.

In 2004, the government announced measures intended to reduce these debts. This included a new FTB Part A supplement, payable as a lump sum of $600 per year per child, indexed to inflation. The lump sum would offset debts that families had due to overpayments, meaning that overpayments would have a much less significant impact, and those families who had not been overpaid would benefit significantly.5 Families who remain on benefits for the entire year also receive these payments, so that they are very progressive and boost the incomes of low-income families quite significantly. The introduction of these lump sums – albeit with significant budgetary costs – appeared to have resolved the overpayment and debt controversy, at least until recently.

Better management of the social welfare system: AKA ‘Robo-debt’

In late 2016, members of the public raised concerns about letters from the Department of Human Services advising that they owed the government significant debts for past income support payments received. This quickly escalated, with the shadow human services minister requesting the auditor general to investigate, and an independent MP asking the Commonwealth Ombudsman to step in after receiving more than 100 complaints about the debt recovery process. The Senate Community Affairs Reference Committee launched a Parliamentary Inquiry, reporting in June 2017. Its report noted that between November 2016 and March 2017 at least 200,000 people were affected, with 20,000 debt letters being sent each week. In total, the number of ‘debt interventions’ increased from 20,000 in 2015–16 to nearly 800,000 in 2016–17 (Senate Community Affairs Reference Committee, 2017).

The issues followed from a 2015 announcement of a computer upgrade for Centrelink, offering better data analytics, real-time data sharing between agencies, and faster, cheaper implementation of policy changes. According to the then Minister for Human Services, ‘This means customers who fail to update their details with us will be less likely to have to repay large debts, and those who willfully act to defraud taxpayers will be caught much more quickly’ (Cowan and Coyne, 2015). These initiatives followed the introduction of new computer systems in the Tax Office, which introduced the Single Touch Payroll (STP) system, streamlining the way employers report payroll and superannuation information.

From 2016, legislation introduced an interest charge on the debts of former social welfare recipients who were unwilling to enter repayment arrangements; brought in international Departure Prohibition Orders for people not in repayment arrangements; and removed the six-year limitation on debt recovery for all social welfare debt.

The Commonwealth Ombudsman’s Report provides a clear summary of the new system:

In July 2016 the Department of Human Services (DHS)–Centrelink launched a new online compliance intervention (OCI) system for raising and recovering debts. The OCI matches the earnings recorded on a customer’s Centrelink record with historical employer-reported income data from the Australian Taxation Office (ATO). Parts of the debt raising process previously done manually by compliance officers within DHS are now done using this automated process. Customers are asked to confirm or update their income using the online system. If the customer does not engage with DHS either online or in person, or if there are gaps in the information provided by the customer, the system will fill the gaps with a fortnightly income figure derived from the ATO income data for the relevant employment period (‘averaged’ data) (Commonwealth Ombudsman, 2017, page 1).

Data-matching between the benefits and tax authorities is not new, but these measures automated the system of identifying anomalies between income reported to the ATO and income reported to the DHS. Hence it was characterised as ‘robo-debt’.6 In addition, whereas previously the DHS collected information from employers, the new system shifted responsibility for providing information on to the individuals concerned.

A significant problem with this system is that when individuals are asked to confirm online their annual income reported to the ATO, the algorithm used by the DHS could produce false estimates of debts by dividing the annual wages reported by 26. This approach will only work correctly if individuals receive exactly the same income each fortnight. For students, for example, this is very unlikely – the beginning and end of the academic year do not coincide with the tax year, so someone who combines part-time work and study, and then graduates and moves into full-time work, is likely to have six months at a much higher wage – and therefore potentially incur a debt if income is averaged.

Similarly, people who move between different jobs, or who have short interruptions to work and claim sickness benefits because they are casuals and not entitled to paid leave, are also potentially exposed to debts under this approach. A significant proportion of Australian workers have variable earnings. In 2018, 24 per cent of all employees reported earnings varying from one period to the next (excluding overtime payments), 21 per cent often worked a different number of hours each week, and 19 per cent had no guaranteed minimum number of hours each week (ABS, 2018). People who move from welfare to work are most exposed to having debts miscalculated.

The number of people for whom the DHS calculations are incorrect is not known. Tracking outcomes is complex, as debts may be issued in one period, but not finalised until another. In approximately 20 per cent of cases in which an individual received an initial letter identifying a discrepancy between ATO and DHS information, the individual was able to provide clarifying information, resulting in no debt owed (O’Donovan, 2019).

By July 2018, it was reported that following requests for a review, around 33,000 debts were settled for less than the amount initially raised, with the overall amount of debt reduced by 75 per cent (O’Donovan, 2019). In the first two years of the scheme only 36 per cent of the debts raised were recovered, and about half of the 300,000 people for whom debts were raised had paid the full debt. This may be an overestimate of the ‘accuracy’ of the system, if people choose to pay the debts claimed rather than go through the technically onerous and time-consuming nature of questioning and appealing the debts (Senate Budget Estimates, 2019).

A recurring feature of these problems is a lack of transparency about the use of averaging procedures and the extent of mistakes leading to ‘over-recovery’. Indeed, in an answer to a Senate Question on Notice the DHS said that the number of people for whom income was averaged to calculate debts ‘was not readily available’ (Senate Additional Estimates, 2019). The Ombudsman’s Report noted that the DHS had not undertaken any modelling of ‘over-recovery’ of debts. In May 2019, the DHS completely dropped a debt against an individual, which, if tested in a court of competent Commonwealth jurisdiction, may have led to the procedures being found to be unlawful (Henriques-Gomez, 2019).

There are a number of other important policy issues. The shift of responsibility from the DHS on to individuals to collect their earnings details, sometimes for periods six or seven years earlier, has real economic costs, due to the time required to gather this information, exacerbated by the difficulty in contacting Centrelink by telephone.

There are serious legal questions about the new procedure. Carney (2019) has argued that the failure of a person to disprove the possibility of a debt is not a legal foundation for a debt and is inconsistent with the principles of sound administration. O’Donovan (2019) also notes the unfair impact of the combination of the reversal of onus of proof, income averaging and the system’s lack of transparency:

The word ‘integrity’ and ‘right payment to the right person’ often get thrown around as justifying this programme. Firstly the department has never published any analysis of whether debts are over or under calculated, despite an Ombudsman ‘suggestion’ it do so. Regardless of your political belief, debts should not be determined by a person’s capacity to contest and the financial and time resources available to them. But that is what we now have.

The controversy is ongoing. In July 2019, the Federal Opposition called for the programme to be dropped (whereas they had previously called for it to be suspended), and in July 2019 the Senate voted for a second Inquiry into the scheme. The Victorian Legal Aid Centre (2019) has mounted a test case against the scheme, and in September 2019 the DHS again completely dropped this debt, although the test case will continue. Also, in September 2019, the Opposition announced that it was supporting a class action against the programme.7

It is possible that these problems will decrease over time. The 2019–20 Federal Budget introduced changes to the assessment model. This involves the ‘real-time’ transmission of earnings data from employers to the DHS. From 1 July 2020, income support recipients who are employed will report income received during the fortnight, with income data shared with the DHS through expanded data-sharing arrangements. That is, earnings assessed will change from what the employer is due to pay to the earnings actually received. This measure will supposedly reduce the likelihood of overpayments and therefore of debts. In addition, real-time reporting will mean that the administrative burden will be shared between employers and recipients.

The design of income tests in social security policy

Designing social protection systems that provide timely support as income and circumstances change raises many challenges. Here we reflect on two issues: the creation of debts by design and the appropriate level of responsiveness.

Debt by design?

Both the UK and Australia have designed income-tested systems that inevitably create overpayments. In part this has been because of decisions to design a system that involves ‘pay now, reconcile later’. In Australia using estimates of prospective annual income lead to under-estimates of income and so overpayments. In the UK using retrospective annual income underestimated the number and extent of changes in the year of the award, again leading to overpayments. Furthermore, the seeking out of old overpayments has been made easier by systems of data matching and automation, and the application of this has led to scenarios like the robo-debt problems.

Overpayments are not necessarily debts, however. As we have seen, these can be reduced or removed in various ways, including by setting disregards to stop overpayments building up, by waivers in full or part after the fact, or by increasing benefits as Australia did in effect by introducing end-of-year lump sums. And/or repayment periods can be slowed down, so that the impact of debt recovery is spread over a longer period. These are policy decisions.

The issues that arise with debts in Universal Credit are different. They reflect conscious policy decisions to make payments monthly in arrears, with an initial significant waiting period, which means that people have to rely on other sources of income or go into debt. The Australian system of income support appears to have avoided some of these problems by having a shorter payment period of a fortnight. Other aspects of the Universal Credit design – such as paying rent direct and not to the landlord – have also increased the likelihood that claimants will fall into debt. The UK has introduced a system of ‘advance’ payments, so that people can receive up to 100 per cent of their estimated Universal Credit entitlement at the start of the claim. But these are loans, and must be repaid.

The reconciliation of incomes and payments is clearly important, and most would agree with the principle that households should receive their correct entitlements. Having said this, in each of these cases overpayments and debts are at least partly the consequence of deliberate policy design choices, and very large numbers of low-income people in both countries suffered hardship as a result. Some of the recovery of old overpayments go back over several years. But the time periods for the backdating of benefits (where claims have been delayed for example) tend to be much shorter, in months rather than years. There is thus an imbalance, or even double standards, at play here (Harris, 2014). Waivers, disregards and advances can reduce debts.8 But it would clearly be preferable to have payment designs that as far as possible avoided this problem in advance.

Security and responsiveness

As noted earlier, the OECD has specifically recommended that means tests should be ‘more responsive to people’s needs by shortening the reference periods for needs assessments’ (2019b: 325). But is greater responsiveness in income-tested systems actually a desirable feature? In theory responsiveness – where benefit payments track other income changes over relatively short time periods – provides a system that is efficient and fair. In practice, there are trade-offs across the time periods involved.

There are three key time periods – the assessment period, the frequency of payment, and the length of time an award is fixed before re-assessment – and there are pros and cons to consider in the choices to be made. Short assessment periods may be atypical of usual earnings, be open to manipulation (no overtime for a few weeks) or collusion (employer agrees to reduced hours for the assessment period). Long assessment periods may provide a more accurate measure of income smoothed over time, being less subject to lumpy payments or spikes. Short frequency of payment periods can provide a regular and stable source of income. Long frequency can make budgeting more difficult for some. Having the amount of an award fixed for short time periods means payments may go up and down before people have time to adjust. Long time periods for awards means that there is a risk that a gap will open up between needs and benefit income, sometimes to the advantage of the claimant, sometimes not. Each of these choices have different implications for simplicity, for the compliance costs for claimants, and for the administrative effort.

Responsiveness must also be traded against security. Security of income – knowing how much you will receive and when – is an important dimension of not just how people manage their money but also how they feel about their financial situation (Ridge and Millar, 2011; Datta, 2019; Summers and Young, 2019). Changing benefits as soon as circumstances change adds to financial insecurity, potentially creating stress alongside debts. And short-term change may also be a poor guide to longer-term need. As Romich and Hill (2017: 3) note,

Ideally, families exit income support programs after a permanent, or sustained, increase in income. Yet, the reality of low-income family life shows that changes in employment and family structure are rarely permanent. Program rules could make a family seem eligible in some periods but not in others, when the family’s needs have not fundamentally changed […]. Most churning is related to changes in residence, employment, and household composition, and many of those changes are short-lived.

The UK and Australian case studies highlight the problems that can arise when overpayments and debts are built into the design of benefit systems and when benefits respond very quickly to changes in earnings and other circumstances. Many citizens have been directly, often adversely, affected by these policy choices, advertently or inadvertently. These issues are far from resolved in these countries, with ongoing legal challenges in both the UK and Australia, as well as political and popular debate about the future purpose and design of income-tested benefits for working-age people. Timing issues are increasingly visible, as income-testing and in-work benefits are expanding. There will always be trade-offs in designing such benefits but there is much that can be learnt from these UK and Australian policy examples.

Acknowledgements

We are grateful for comments from Fran Bennett, Terry Carney, Rita Griffiths, Alexandra Heron and Andrew Sinstead-Reid.

Notes

1

The four income-tested benefits are usually paid two weeks in arrears after a seven-day waiting period. Housing Benefit is usually paid direct to the landlord and the payment frequency depends on when rent is due (so can be weekly, fortnightly, four-weekly or monthly). Most families with children are also eligible for Child Benefit, which is paid every four weeks.

2

From April 2012, there has also been a disregard for falls in income of £2,500. These underpayments tend to attract less policy attention than the overpayments, but around 800,000 people were underpaid in 2016/7, to the tune of about £550 million, with possible financial hardship as a result (HMRC, 2018). Underpayments are repaid as lump sums, usually automatically after the year-end reconciliation.

3

In addition to separate payments for different categories (unemployed people, lone parents, carers and people with disability), payments for children and childcare support are administered separately.

4

On the other hand, around 271,000 families received additional assistance – an average of $1,028 per family – because they had initially overestimated their incomes. This top-up would not have occurred under the previous system.

5

Similarly, there was the introduction of a lump sum payment of around $300 per family for Family Tax Benefit Part B for single earner families. The payments were indexed, and since 2004 have increased in nominal terms from $600 to $766.50 per child (FTBA) and from just over $300 to $372.30 per family for single earner families (FTBB).

6

In the USA, the term ‘zombie debt’ has been applied to the identification and collection of old overpayments of benefits, sometimes going back decades, and also causing stress and hardship for some (Eubanks, 2019).

7

From November 2019 some “refinements” to the income compliance system have been introduced (Robert, 2019). At this stage, it appears that the proposed class action and another legal case will proceed, so that the future of the system remains unclear.

8

See also the National Audit Office (2018) report on the need for further government action to help to reduce and manage ‘problem debt’.

Conflict of interest

The authors declare that there is no conflict of interest.

References

  • ABS (Australian Bureau of Statistics) (2018) Survey of Characteristics of Employment, Catalogue No. 63333.0, Canberra: ABS.

  • Barr, N. (1987) The Economics of the Welfare State, London: Weidenfeld and Nicholson.

  • Campbell, I. and Burgess, J. (2018) Patchy progress? Two decades of research on precariousness and precarious work in Australia, Labour & Industry: A Journal of the Social and Economic Relations of Work, 1(28): 4867.

    • Search Google Scholar
    • Export Citation
  • Carney, T. (2019 ) Robo-debt illegality: the seven veils of failed guarantees of the rule of law?, Alternative Law Journal, 44(1): 410, doi: 10.117/1037969X18815913.

    • Search Google Scholar
    • Export Citation
  • Charlesworth, Z. (2018) Three significant but little known welfare changes for April 2018, http://policyinpractice.co.uk/three-significant-but-little-known-welfare-changes-live-from-april-2018.

    • Search Google Scholar
    • Export Citation
  • Chote, R., Emmerson, C., Harrison, R. and Miles, D. (2006) The IFS Green Budget: January 2006, London: Institute for Fiscal Studies.

  • Clasen, J. (2019) Subsidizing wages or supplementing transfers? The politics and ambiguity of in‐work benefits, Social Policy & Administration, 113, doi: 10.1111/spol.12488.

    • Search Google Scholar
    • Export Citation
  • Clegg, D. (2013) Europe, Labour Market Megatrends and Social Security, Geneva: International Social Security Association, pp 4756.

  • Cowan, P. and Coyne, A. (2015) Cabinet approves $1bn centrelink systems upgrade, IT News, www.itnews.com.au/news/cabinet-approves-1bn-centrelink-systems-upgrade-402586.

    • Search Google Scholar
    • Export Citation
  • Datta, N. (2019) Willing to pay for security: a discrete choice experiment to analyse labour supply preferences, CEP Discussion Paper No 1632, London: Centre for Economic Performance, London School of Economics and Political Science.

    • Search Google Scholar
    • Export Citation
  • DSS (Department of Social Services) (2019) DSS payment demographic data, March, 2019, Greenway, ACT: DSS.

  • DWP (Department for Work and Pensions) (2018) Universal Credit Full Service Survey, London: DWP.

  • DWP (Department for Work and Pensions) (2019) Universal Credit Statistics, London: DWP.

  • Eubanks, V. (2019) Zombie debts are hounding struggling Americans. Will you be next?, The Guardian [Online], www.theguardian.com/law/2019/oct/15/zombie-debt-benefits-overpayment-poverty.

    • Search Google Scholar
    • Export Citation
  • Greve, B. (2019) The digital economy and the future of European welfare states, International Social Security Review, 72: 7994, doi: 10.1111/issr.12214.

    • Search Google Scholar
    • Export Citation
  • Hardy, G. and Lane, J. (2018) Walking on Thin Ice: The Cost of Financial Insecurity, Oxford: Citizens Advice.

  • Harris, N. (2014) Law in a Complex State: Complexity in the Law and Structure of Welfare, London: Hart Publishing.

  • House of Commons Committee of Public Accounts (2018) HMRC’s Performance in 2017–18 Sixty-Sixth Report of Session 2017–19, London: The Stationery Office.

    • Search Google Scholar
    • Export Citation
  • House of Commons Treasury Select Committee (2006) The Administration of Tax Credits: Sixth Report of Session 2005–06, HC 49, London: The Stationery Office.

    • Search Google Scholar
    • Export Citation
  • House of Commons Work and Pensions Select Committee (2018) Universal Credit Childcare, London: The Stationery Office.

  • HMRC (2019) Child and Working Tax Credits, Finalised Awards, 2017–2018, London: HMRC.

  • Hills, J., Smithies, R. and McKnight, A. (2006) Tracking Income: How Working Families’ Incomes Vary Through the Year, London: CASEreport.

    • Search Google Scholar
    • Export Citation
  • Henriques-Gomez, L. (2019) Centrelink drops woman’s robodebt after she mounts court challenge, www.theguardian.com/australia-news/2019/may/05/centrelink-drops-womans-robodebt-after-she-mounts-court-challenge.

    • Search Google Scholar
    • Export Citation
  • IMF (International Monetary Fund) (2018) G-20 Background Note, The Future of Work: Measurement and Policy Challenges, Washington, DC: IMF (18 July 2018).

    • Search Google Scholar
    • Export Citation
  • JItendra, A., Thompson, E., Rabindrakumar, S., Brownfield, G., Butler, A. and Tutton, P. (2019) Hardship Now or hardship Later: Universal Credit, Debt and the Five Week Wait, London: Trussell Trust.

    • Search Google Scholar
    • Export Citation
  • Kennedy, S. and Keen, R. (2018) Universal Credit Roll-Out: 2018–19, House of Commons Library, Briefing Paper Number 8299, 14 June 2018.

    • Search Google Scholar
    • Export Citation
  • Kennedy, S. and Mackley, A. (2019) The Universal Credit Assessment Period and Earned Income, House of Commons Library, Briefing Paper Number 8501.

    • Search Google Scholar
    • Export Citation
  • Kennedy, S., Mackley, A., Wilson, W. and O’Donnell, M. (2019) Universal Credit and Debt, House of Commons Library, Debate Pack number CDP-2019-0131.

    • Search Google Scholar
    • Export Citation
  • Lords Hansard (2006) Tax Credits Debate, London: Houses of Parliament, (23 October).

  • McKinnon, R. (2019) Introduction: social security and the digital economy – managing transformation, International Social Security Review, 72: 516, doi: 10.1111/issr.12211.

    • Search Google Scholar
    • Export Citation
  • Millar, J. (2008) Making work pay, making tax credits work: an assessment with specific reference to lone-parent employment, International Social Security Review, 61(2): 2138. doi: 10.1111/j.1468-246X.2008.00308.x

    • Search Google Scholar
    • Export Citation
  • Millar, J. and Bennett, F. (2017) Universal credit: assumptions, contradictions and virtual reality, Social Policy and Society, 16(2): 16982. doi: 10.1017/S1474746416000154

    • Search Google Scholar
    • Export Citation
  • Millar, J. and Whiteford, P. (2017) Timing it wrong: benefits, income tests, overpayments and debts, Inside Story, 27 February. http://blogs.bath.ac.uk/iprblog/2017/03/01/timing-it-wrong-benefits-income-tests-overpayments-and-debts/

    • Search Google Scholar
    • Export Citation
  • Morduch, J. and Schneider, R. (2017) The Financial Diaries, Princeton, NJ: Princeton University Press.

  • National Audit Office (2007) HM Revenue & Customs 2006–7 Accounts, the Comptroller and Auditor General’s Standard Report, London: Comptroller and Auditor General.

    • Search Google Scholar
    • Export Citation
  • National Audit Office (2018) Tackling Problem Debt, London: Comptroller and Auditor General.

  • National Audit Office (2019) Department for Work and Pensions Department Overview, London: Comptroller and Auditor General.

  • O’Donovan, D. (2019) ‘Dial 1800 – Reverse Onus’: coming to grips with robodebt, https://law.blogs.latrobe.edu.au/.

  • OECD (2016) Social Expenditure Update 2016, Paris: OECD,www.oecd.org/els/soc/OECD2016-Social-Expenditure-Update.pdf.

  • OECD (2019a) Policy Responses to New Forms of Work, Paris: OECD.

  • OECD (2019b) Employment Outlook 2019, Paris: OECD.

  • ONS (2018) Contracts that do not Guarantee a Minimum Number of Hours: April 2018, London: ONS.

  • Parliamentary Ombudsman (2007) Tax Credits: Getting it wrong? 5th Report Session 2006–2007, HC 1010, London: House of Commons.

  • Revenue Benefits (2019) How do tax credits work? Understanding the disregard, https://revenuebenefits.org.uk/tax-credits/guidance/how-do-tax-credits-work/understanding-the-disregard/#History%20of%20the%20income%20disregards.

    • Search Google Scholar
    • Export Citation
  • Ridge, T. and Millar, J. (2011) Following families: working lone‐mother families and their children, Social Policy & Administration, 45(1): 8597.

    • Search Google Scholar
    • Export Citation
  • Robert, S. (2019) Transcript: ABC News, Doorstop, topics: Centrelink’s online compliance system https://www.mhs.gov.au/transcripts/2019-11-19-transcript-abc-news-doorstop

    • Search Google Scholar
    • Export Citation
  • Romich, J. and Hill, H.D. (2017) Income instability and income support programs: recommendations for policy and practice, the consortium, www.opressrc.org/sites/default/files/public/IB_Income%20Instability_082917.pdf.

    • Search Google Scholar
    • Export Citation
  • Rubery, J., Grimshaw, D., Keizer, A. and Johnson, M. (2018) Challenges and contradictions in the ‘normalising’ of precarious work, Work, Employment and Society, 32(3): 50927. doi: 10.1177/0950017017751790

    • Search Google Scholar
    • Export Citation
  • Senate Community Affairs Reference Committee, Australia, (2017) Design, scope, cost-benefit analysis, contracts awarded and implementation associated with the Better Management of the Social Welfare System initiative, Commonwealth of Australia, Canberra.

    • Search Google Scholar
    • Export Citation
  • Senate Community Affairs Legislation Committee Australia, (2019) Budget estimates 2019–20 – 5 April 2019 answer to question on notice, Department of Human Services, Topic: Online Compliance Intervention – South Australia, Question reference number: HS 7 (SQ19-000113), Senate Hansard, Canberra.

    • Search Google Scholar
    • Export Citation
  • Senate Community Affairs Legislation Committee, Australia, (2019) Additional estimates 2018–19 – 21 February 2019 answer to question on notice, Department of Human Services, Topic: Online Compliance Intervention – Debt Repayments Without Review, Question reference number: HS 26 (SQ19-000039), Senate Hansard, Canberra.

    • Search Google Scholar
    • Export Citation
  • SSAC (Social Security Advisory Committee) (2018) The Draft Universal Credit (Managed Migration) Regulations 2018, London: Stationery Office.

    • Search Google Scholar
    • Export Citation
  • Stewart, M. and Whiteford, P. (2018) Balancing equity and efficiency in the tax and transfer system, in R. Breunig and M. Fabian (eds) Hybrid Public Policy Innovations: Contemporary Policy Beyond Ideology, Abingdon: Routledge.

    • Search Google Scholar
    • Export Citation
  • Summers, K. and Young, D. (2019) The alleged simplicity of Universal Credit and the lived experience of benefit claimants, https://blogs.lse.ac.uk/politicsandpolicy/universal-credit-simplicity/.

    • Search Google Scholar
    • Export Citation
  • Tomlinson, D. (2018) Irregular Payments: Assessing the Breadth and Depth of Month to Month Earnings Volatility, London: The Resolution Foundation.

    • Search Google Scholar
    • Export Citation
  • Todd, V. (2018) Real time information and benefits: an update, Tax Adviser, May. https://www.taxadvisermagazine.com/article/real-time-information-and-benefits-update

    • Search Google Scholar
    • Export Citation
  • Tucker, J. (2018) Universal Credit: What Needs to Change to Reduce Child Poverty and Make it Fit for Families?, London: CPAG.

  • Tucker, J. and Norris, D. (2018) Rough Justice. Problems with Monthly Assessment of Pay and Circumstances in Universal Credit, and What can be Done About Them, London: CPAG.

    • Search Google Scholar
    • Export Citation
  • Victoria Legal Aid (2019) Centrelink waives another robo-debt as legal challenge continues, www.legalaid.vic.gov.au/about-us/news/centrelink-waives-another-robo-debt-legal-challenge-continues.

    • Search Google Scholar
    • Export Citation
  • Whiteford, P., Mendelson, M. and Millar, J. (2003) Timing it Right? Tax Credits and how to Respond to Income Changes, York: Joseph Rowntree Foundation.

    • Search Google Scholar
    • Export Citation
  • World Bank (2019) World Development Report 2019: The Changing Nature of Work, Washington, DC: World Bank Group.

  • ABS (Australian Bureau of Statistics) (2018) Survey of Characteristics of Employment, Catalogue No. 63333.0, Canberra: ABS.

  • Barr, N. (1987) The Economics of the Welfare State, London: Weidenfeld and Nicholson.

  • Campbell, I. and Burgess, J. (2018) Patchy progress? Two decades of research on precariousness and precarious work in Australia, Labour & Industry: A Journal of the Social and Economic Relations of Work, 1(28): 4867.

    • Search Google Scholar
    • Export Citation
  • Carney, T. (2019 ) Robo-debt illegality: the seven veils of failed guarantees of the rule of law?, Alternative Law Journal, 44(1): 410, doi: 10.117/1037969X18815913.

    • Search Google Scholar
    • Export Citation
  • Charlesworth, Z. (2018) Three significant but little known welfare changes for April 2018, http://policyinpractice.co.uk/three-significant-but-little-known-welfare-changes-live-from-april-2018.

    • Search Google Scholar
    • Export Citation
  • Chote, R., Emmerson, C., Harrison, R. and Miles, D. (2006) The IFS Green Budget: January 2006, London: Institute for Fiscal Studies.

  • Clasen, J. (2019) Subsidizing wages or supplementing transfers? The politics and ambiguity of in‐work benefits, Social Policy & Administration, 113, doi: 10.1111/spol.12488.

    • Search Google Scholar
    • Export Citation
  • Clegg, D. (2013) Europe, Labour Market Megatrends and Social Security, Geneva: International Social Security Association, pp 4756.

  • Cowan, P. and Coyne, A. (2015) Cabinet approves $1bn centrelink systems upgrade, IT News, www.itnews.com.au/news/cabinet-approves-1bn-centrelink-systems-upgrade-402586.

    • Search Google Scholar
    • Export Citation
  • Datta, N. (2019) Willing to pay for security: a discrete choice experiment to analyse labour supply preferences, CEP Discussion Paper No 1632, London: Centre for Economic Performance, London School of Economics and Political Science.

    • Search Google Scholar
    • Export Citation
  • DSS (Department of Social Services) (2019) DSS payment demographic data, March, 2019, Greenway, ACT: DSS.

  • DWP (Department for Work and Pensions) (2018) Universal Credit Full Service Survey, London: DWP.

  • DWP (Department for Work and Pensions) (2019) Universal Credit Statistics, London: DWP.

  • Eubanks, V. (2019) Zombie debts are hounding struggling Americans. Will you be next?, The Guardian [Online], www.theguardian.com/law/2019/oct/15/zombie-debt-benefits-overpayment-poverty.

    • Search Google Scholar
    • Export Citation
  • Greve, B. (2019) The digital economy and the future of European welfare states, International Social Security Review, 72: 7994, doi: 10.1111/issr.12214.

    • Search Google Scholar
    • Export Citation
  • Hardy, G. and Lane, J. (2018) Walking on Thin Ice: The Cost of Financial Insecurity, Oxford: Citizens Advice.

  • Harris, N. (2014) Law in a Complex State: Complexity in the Law and Structure of Welfare, London: Hart Publishing.

  • House of Commons Committee of Public Accounts (2018) HMRC’s Performance in 2017–18 Sixty-Sixth Report of Session 2017–19, London: The Stationery Office.

    • Search Google Scholar
    • Export Citation
  • House of Commons Treasury Select Committee (2006) The Administration of Tax Credits: Sixth Report of Session 2005–06, HC 49, London: The Stationery Office.

    • Search Google Scholar
    • Export Citation
  • House of Commons Work and Pensions Select Committee (2018) Universal Credit Childcare, London: The Stationery Office.

  • HMRC (2019) Child and Working Tax Credits, Finalised Awards, 2017–2018, London: HMRC.

  • Hills, J., Smithies, R. and McKnight, A. (2006) Tracking Income: How Working Families’ Incomes Vary Through the Year, London: CASEreport.

    • Search Google Scholar
    • Export Citation
  • Henriques-Gomez, L. (2019) Centrelink drops woman’s robodebt after she mounts court challenge, www.theguardian.com/australia-news/2019/may/05/centrelink-drops-womans-robodebt-after-she-mounts-court-challenge.

    • Search Google Scholar
    • Export Citation
  • IMF (International Monetary Fund) (2018) G-20 Background Note, The Future of Work: Measurement and Policy Challenges, Washington, DC: IMF (18 July 2018).

    • Search Google Scholar
    • Export Citation
  • JItendra, A., Thompson, E., Rabindrakumar, S., Brownfield, G., Butler, A. and Tutton, P. (2019) Hardship Now or hardship Later: Universal Credit, Debt and the Five Week Wait, London: Trussell Trust.

    • Search Google Scholar
    • Export Citation
  • Kennedy, S. and Keen, R. (2018) Universal Credit Roll-Out: 2018–19, House of Commons Library, Briefing Paper Number 8299, 14 June 2018.

    • Search Google Scholar
    • Export Citation
  • Kennedy, S. and Mackley, A. (2019) The Universal Credit Assessment Period and Earned Income, House of Commons Library, Briefing Paper Number 8501.

    • Search Google Scholar
    • Export Citation
  • Kennedy, S., Mackley, A., Wilson, W. and O’Donnell, M. (2019) Universal Credit and Debt, House of Commons Library, Debate Pack number CDP-2019-0131.

    • Search Google Scholar
    • Export Citation
  • Lords Hansard (2006) Tax Credits Debate, London: Houses of Parliament, (23 October).

  • McKinnon, R. (2019) Introduction: social security and the digital economy – managing transformation, International Social Security Review, 72: 516, doi: 10.1111/issr.12211.

    • Search Google Scholar
    • Export Citation
  • Millar, J. (2008) Making work pay, making tax credits work: an assessment with specific reference to lone-parent employment, International Social Security Review, 61(2): 2138. doi: 10.1111/j.1468-246X.2008.00308.x

    • Search Google Scholar
    • Export Citation
  • Millar, J. and Bennett, F. (2017) Universal credit: assumptions, contradictions and virtual reality, Social Policy and Society, 16(2): 16982. doi: 10.1017/S1474746416000154

    • Search Google Scholar
    • Export Citation
  • Millar, J. and Whiteford, P. (2017) Timing it wrong: benefits, income tests, overpayments and debts, Inside Story, 27 February. http://blogs.bath.ac.uk/iprblog/2017/03/01/timing-it-wrong-benefits-income-tests-overpayments-and-debts/

    • Search Google Scholar
    • Export Citation
  • Morduch, J. and Schneider, R. (2017) The Financial Diaries, Princeton, NJ: Princeton University Press.

  • National Audit Office (2007) HM Revenue & Customs 2006–7 Accounts, the Comptroller and Auditor General’s Standard Report, London: Comptroller and Auditor General.

    • Search Google Scholar
    • Export Citation
  • National Audit Office (2018) Tackling Problem Debt, London: Comptroller and Auditor General.

  • National Audit Office (2019) Department for Work and Pensions Department Overview, London: Comptroller and Auditor General.

  • O’Donovan, D. (2019) ‘Dial 1800 – Reverse Onus’: coming to grips with robodebt, https://law.blogs.latrobe.edu.au/.

  • OECD (2016) Social Expenditure Update 2016, Paris: OECD,www.oecd.org/els/soc/OECD2016-Social-Expenditure-Update.pdf.

  • OECD (2019a) Policy Responses to New Forms of Work, Paris: OECD.

  • OECD (2019b) Employment Outlook 2019, Paris: OECD.

  • ONS (2018) Contracts that do not Guarantee a Minimum Number of Hours: April 2018, London: ONS.

  • Parliamentary Ombudsman (2007) Tax Credits: Getting it wrong? 5th Report Session 2006–2007, HC 1010, London: House of Commons.

  • Revenue Benefits (2019) How do tax credits work? Understanding the disregard, https://revenuebenefits.org.uk/tax-credits/guidance/how-do-tax-credits-work/understanding-the-disregard/#History%20of%20the%20income%20disregards.

    • Search Google Scholar
    • Export Citation
  • Ridge, T. and Millar, J. (2011) Following families: working lone‐mother families and their children, Social Policy & Administration, 45(1): 8597.

    • Search Google Scholar
    • Export Citation
  • Robert, S. (2019) Transcript: ABC News, Doorstop, topics: Centrelink’s online compliance system https://www.mhs.gov.au/transcripts/2019-11-19-transcript-abc-news-doorstop

    • Search Google Scholar
    • Export Citation
  • Romich, J. and Hill, H.D. (2017) Income instability and income support programs: recommendations for policy and practice, the consortium, www.opressrc.org/sites/default/files/public/IB_Income%20Instability_082917.pdf.

    • Search Google Scholar
    • Export Citation
  • Rubery, J., Grimshaw, D., Keizer, A. and Johnson, M. (2018) Challenges and contradictions in the ‘normalising’ of precarious work, Work, Employment and Society, 32(3): 50927. doi: 10.1177/0950017017751790

    • Search Google Scholar
    • Export Citation
  • Senate Community Affairs Reference Committee, Australia, (2017) Design, scope, cost-benefit analysis, contracts awarded and implementation associated with the Better Management of the Social Welfare System initiative, Commonwealth of Australia, Canberra.

    • Search Google Scholar
    • Export Citation
  • Senate Community Affairs Legislation Committee Australia, (2019) Budget estimates 2019–20 – 5 April 2019 answer to question on notice, Department of Human Services, Topic: Online Compliance Intervention – South Australia, Question reference number: HS 7 (SQ19-000113), Senate Hansard, Canberra.

    • Search Google Scholar
    • Export Citation
  • Senate Community Affairs Legislation Committee, Australia, (2019) Additional estimates 2018–19 – 21 February 2019 answer to question on notice, Department of Human Services, Topic: Online Compliance Intervention – Debt Repayments Without Review, Question reference number: HS 26 (SQ19-000039), Senate Hansard, Canberra.

    • Search Google Scholar
    • Export Citation
  • SSAC (Social Security Advisory Committee) (2018) The Draft Universal Credit (Managed Migration) Regulations 2018, London: Stationery Office.

    • Search Google Scholar
    • Export Citation
  • Stewart, M. and Whiteford, P. (2018) Balancing equity and efficiency in the tax and transfer system, in R. Breunig and M. Fabian (eds) Hybrid Public Policy Innovations: Contemporary Policy Beyond Ideology, Abingdon: Routledge.

    • Search Google Scholar
    • Export Citation
  • Summers, K. and Young, D. (2019) The alleged simplicity of Universal Credit and the lived experience of benefit claimants, https://blogs.lse.ac.uk/politicsandpolicy/universal-credit-simplicity/.

    • Search Google Scholar
    • Export Citation
  • Tomlinson, D. (2018) Irregular Payments: Assessing the Breadth and Depth of Month to Month Earnings Volatility, London: The Resolution Foundation.

    • Search Google Scholar
    • Export Citation
  • Todd, V. (2018) Real time information and benefits: an update, Tax Adviser, May. https://www.taxadvisermagazine.com/article/real-time-information-and-benefits-update

    • Search Google Scholar
    • Export Citation
  • Tucker, J. (2018) Universal Credit: What Needs to Change to Reduce Child Poverty and Make it Fit for Families?, London: CPAG.

  • Tucker, J. and Norris, D. (2018) Rough Justice. Problems with Monthly Assessment of Pay and Circumstances in Universal Credit, and What can be Done About Them, London: CPAG.

    • Search Google Scholar
    • Export Citation
  • Victoria Legal Aid (2019) Centrelink waives another robo-debt as legal challenge continues, www.legalaid.vic.gov.au/about-us/news/centrelink-waives-another-robo-debt-legal-challenge-continues.

    • Search Google Scholar
    • Export Citation
  • Whiteford, P., Mendelson, M. and Millar, J. (2003) Timing it Right? Tax Credits and how to Respond to Income Changes, York: Joseph Rowntree Foundation.

    • Search Google Scholar
    • Export Citation
  • World Bank (2019) World Development Report 2019: The Changing Nature of Work, Washington, DC: World Bank Group.

  • 1 University of Bath, , UK
  • | 2 Australian National University, , Australia

Content Metrics

May 2022 onwards Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 19 19 19
PDF Downloads 13 13 13

Altmetrics

Dimensions