What do we know about In-Work Poverty? A summary of the evidence

2 Summary of the Evidence

Extent and impact of in-work poverty

Moving into work doesn't necessarily mean moving out of poverty

2.1 There is a strong relationship between worklessness and poverty, both in the UK and internationally. Poverty rates^[1] for households with all adults in employment are considerably lower than in workless households.

Chart 1 - Risk of poverty by household employment status, Scotland

Source: HBAI dataset, DWP

2.2 However, employment by no means guarantees a route out of poverty. The majority (52%) of working age adults in poverty^[2] in Scotland are in 'in-work poverty' i.e. they are living in households with at least one adult in employment. For children in poverty, 59% are living in households with someone in employment.^[3] In 31% of households in the bottom three income deciles, at least one adult is working full time.^[4]

The proportion of those in in-work poverty is gradually increasing

Chart 2 - Percentage of children and working age adults in poverty who are in in-work poverty^[5]

Source: HBAI dataset, DWP

2.3 The nature of poverty in Scotland, and in the UK, has changed over the last decade. Over this period, there have been significant decreases in the rates of relative poverty for both working age adults and children, but without corresponding decreases in the rate of in-work poverty for working age adults and children.

2.4 The result is that the percentages of children and working age adults in poverty who are in in-work poverty have increased gradually over time.

In-work poverty is a problem for individuals and society

2.5 The experience of living in poverty can impact on many areas of life. It goes beyond having to live on a low income and can affect access to services, right to participation in society, health, educational attainment and overall life chances. Growing up in poverty can have a profound and lasting impact on children's outcomes - income poverty and material deprivation are strongly associated with poorer outcomes for children. This is not simply an issue of exclusion experienced as a direct result of a lack of material resources, but with a range of interconnected issues, such as stress and poor health.^[6]

2.6 In-work poverty is also costly to society, both via payment of tax credits and other in-work benefits which top up low income, and through foregone tax revenues. The long-term costs of allowing 110,000 children in Scotland to be affected by in-work poverty in 2012/13^[7] could be around £1.1 billion^[8].

Factors influencing in-work poverty - pay, work intensity, and the welfare and tax systems

2.7 There are three main, inter-related factors that influence the amount of income a household receives, all of which need to be targeted as part of a strategy to tackle in-work poverty: the hourly rate of pay, the number of hours worked ('work intensity'), and income gained and lost through the welfare and tax systems.

2.8 Poverty risk also varies by household composition, with households with children faring worse than their childless counterparts with comparable employment status. This is partly due to the fact that households of different sizes and compositions require a different level of income to achieve the same standard of living.^[9]

Low pay is a key driver of in-work poverty

2.9 Low paid workers face a much higher risk of poverty than workers who are not low paid. UK-level estimates are that low-paid^[10], full-time, year-round employees have approximately double the poverty risk of all workers[11], while all low-paid workers, including those working short hours, have three to four times the rate of poverty of all workers^[12]. Having a low hourly wage is a stronger predictor of poverty than working part-time. However, the relationship between pay and poverty is complex.

Low pay does not always result in poverty, and poverty among working families is not always the result of low pay

2.10 While labour market statistics (and interventions) look at individuals, poverty is generally experienced and measured at a household level. The assumption is that, although individuals are the recipients of earnings, those earnings go into a 'household pot', together with income from a range of other sources (although it should be noted that looking only at overall income hides who has power over resources within a family). This means that low pay does not always result in poverty. Clearly, if a low paid-person lives with a much higher-paid person, they are unlikely to live in poverty. Indeed, only 54% of low paid workers live in low to middle-income households.^[13] And poverty among working families is not always the result of low pay. At the UK level, 44% of adults in working poverty in the UK are not paid below the living wage and do not live with anyone who is^[14]. This has strong implications for designing anti-poverty policies. Any measure that targets only individual low-wage workers leaves many of the working poor uncovered.

For non-low-paid households in working poverty, it is the 'work intensity' that seems to matter

2.11 Work intensity is the number of hours of paid work done by household members. Low pay is a stronger predictor of poverty than low hours, but both are important.

2.12 In nearly half of non-low-paid households (i.e. households in which all of the working adults earn more than around £7.40 an hour) in working poverty, one person works and one does not, and a further quarter are in households that only have part time workers (and may have a non-working partner).^[15] Having only one earner in the household has become a poverty risk in an era in which the average living standard, and hence the relative poverty threshold, is increasingly determined by the living standard of double-earner households.^[16]

2.13 Rates of in-work poverty are higher in families that only have part time workers (25%) and single-earner couple families in which the sole earner works full time (11%), than in one-and-a-half earner families (5%) and households where all working age adults are in full-time work (2%) (see Chart 1). Indeed, the largest group of families with children living in poverty are 'single breadwinner couple' families.^[18]

Many working poor families could not be expected to work more hours

2.14 The UK Government's expectation on parental employment, which it sets out in its Universal Credit guidance, is that:

• One adult in a couple is in full-time work.
• Lone parents, or the second adult in a couple, work full-time if the youngest child is aged 13+; work part-time if the youngest child is aged 5-12, and are not expected to work if the youngest child is under 5 years old.^[19]

2.15 Based on these expectations, analysis has estimated that 42% of children in working poor households live with adults who might be considered to "have potential to work more hours" under Universal Credit. This assumes the work is available, and also that there are no other barriers to exploiting their work potential, including disability or caring responsibilities.^[20]

Many working households rely on welfare to supplement their income

2.16 Whilst comparatively small numbers of low-paid workers can avoid poverty by their own wages alone, most can avoid poverty through a combination of their own and others' wages and/or benefits and tax credits.^[21] Tax credits are an important source of income for households in low paid work, and can help address in-work poverty. Low paid women are less likely than low-paid men to be able to avoid poverty by their own wages alone (as they are more likely to work part-time, due to caring responsibilities), and are consequently more reliant on either a partner's earnings or on tax credits and benefits to lift them out of poverty. This makes women particularly vulnerable to poverty if family relationships break down. Low paid lone parents are very heavily reliant on tax credits and other benefits to lift them out of poverty.^[22] Wage supplements have a positive impact on poverty reduction, since they are family means-tested and thus well targeted on poor households. Evidence for in-work benefits in the UK (Working Families' Tax Credit) and the US shows positive outcomes on poverty reduction.

Pay

Low pay is a particular problem for certain groups

2.17 Those most likely to receive low pay are women, young people, older workers, people without qualifications, some ethnic minorities, lone parents and disabled people. Women are more likely to work in low-paying sectors and are more likely to be employed part-time, which has a substantial overlap with low pay. In fact, it has been argued that low pay is partly a reflection of the under-valuation of women's work, in particular caring work^[23]. There is a big overlap in particular between jobs with high incidence of low pay and the work that many mothers - especially those at high risk of poverty - end up doing when they return to the labour market after having children. In April 2014, the difference between full-time female and male hourly median earnings (excluding overtime) in Scotland was 9.0% (19.8% in the private sector, compared to 0.0% in the public sector)^[24].

2.18 Young people are generally paid significantly less than those over the age of 19^[25], and disabled people and lone parents are more likely to be represented in low-paid jobs.^[26]

2.19 White British men have higher hourly pay than nearly all other employee groups. Some groups have similar hourly wages after controlling for differences in age, occupation and qualifications. However, women, disabled people and members of ethnic minorities are likely to be lower paid and in lower status jobs even with equivalent qualifications. Experimental studies e.g. mock job applications with minority ethnic names or listing disabilities) have shown a causal link with discrimination.^[27]

Low pay is a feature both of the worker and of the job

2.20 The highest qualification held by an individual is important in determining their risk of low pay. Three in ten of those on low pay do not have officially-recognised qualifications; a further one in five is educated to GCSE level.^[28] Research by the Department of Business, Innovation and Skills has shown that on average, achieving a Level 1 qualification adds about 10% to the earnings of a worker with no qualifications.

2.21 However, UK-level analysis has found that, even when qualification level, gender and age are held constant, the incidence of low pay varies significantly by occupation and industrial sector.^[29] A recent survey^[30] of women aged 18+ in the UK, earning £7.44 per hour or less, found that 22% of them had degree level qualifications, and 37% of the total sample described themselves as 'overqualified and overskilled' for their current job. This did not vary significantly between those with and without children. By far the most common reason given for being overqualified was that there were 'no appropriate jobs available'. A focus on increasing skills and qualification levels without a corresponding increase in demand for skills by employers may simply result in people working in jobs they are over-qualified for. The UK has one of the highest reported incidences of the under-use of skills in the workplace in Europe, with self-reported under-use of skills standing at 40%. This compares to 30% in France, 27% in Germany, 24% in Finland and 17% in Austria.^[31]

2.22 Prevalence of low pay is highest among workers in lower- and middle-skilled occupations such as elementary occupations (including cleaners, security guards, catering assistants and leisure workers), sales & customer services (retail assistant, cashiers, telephone salespersons and customer services for example) and personal services (covering social care and childcare for instance). Low pay is much more prevalent in the private sector than in the public or third sectors, and is most likely to occur among those who work in the hotels & restaurants sector, wholesale & retail and admin & support services. The risk of low pay is also higher among workers on temporary contracts,^[32] and those in small or micro-firms compared to those in larger firms.^[33]

2.23 Households in which one or more earners are self-employed have an above-average risk of poverty (22%, compared to 16% for the population as a whole)^[34]. Data regarding poverty rates for self-employed people are thought to be less reliable than for employees, however, due to under-reporting of income and income fluctuations from one year to the next. At the UK level, poverty transitions data show comparatively high rates of poverty entry and exit for self-employed households, which suggest a high level of volatility in income. UK data also shows that comparatively high levels of income poverty among the self-employed are not necessarily reflected in other non-income-based measures of living standards. Although income is difficult to measure reliably for the self-employed, it is known that self-employed people are found at both ends of the labour market - both in higher paid work but also in very low paid work. The recession and subsequent recovery have seen rapid rises in self-employment, at the same time as incomes from self-employment have fallen significantly.^[35]

Scotland (like the rest of the UK) is below average in terms of pay equality

2.24 Compared to other developed countries, Scotland has a relatively large proportion of people working in low-paid jobs. On the basis of the OECD's relative low pay threshold (two-thirds of gross median hourly earnings for full-time employees), 19.5% of employees in Scotland were low paid in 2011: almost three percentage points higher than the OECD average (but lower than the 20.6% for the UK as a whole).

Chart 3 - Share of full-time employees in low paid work (less than two thirds of median full-time pay) across advanced economies, 2011

Source: Annual Survey of Hours and Earnings 2011, ONS; OECD Employment and Labour Market Statistics

2.25 In 2014, 18.4% of employees in Scotland earned less than the living wage (compared to 20.6% in the UK as a whole).^[36] Whilst recent data has shown the first modest increase in real wages since September 2009 and the hope is that this momentum can be maintained, OBR and Bank of England forecasts suggest it may be towards the end of this decade before UK earnings recover to pre-crisis levels.^[37]

Scotland (like the rest of the UK) is below average in terms of productivity

Chart 4 - GDP per hour worked in OECD countries

Source: OECD Productivity Statistics and Scottish Government National Accounts Programme

2.26 Scotland currently matches UK productivity (in terms of GDP per hour worked). However, comparisons to international competitors show Scotland's performance is less favourable. Scotland typically ranks behind more than half of other OECD countries.

2.27 Increasing productivity is crucial to tackling low pay, to enable employers to afford to pay higher wages. However, it is also important to ensure that a reasonable proportion of the productivity gains are passed on to workers in the form of higher wages, as studies show that this is not always the case^[38].

2.28 Changes in the structure of the economy can potentially improve the number of well-paid jobs in the economy - for example, through a strengthening of the manufacturing sector, where earnings tend to be above the national average^[39].

Low paid workers are more likely to become unemployed

2.29 Low-wage workers are less likely to stay in employment than employees on higher wages. This results in what has become termed the 'low-pay, no-pay cycle'. A key factor driving this is the insecurity of low-paid work.^[40] Staying in work can be particularly difficult for low-wage workers with a disability and among individuals who develop a disability or health condition while at work.^[41]

2.30 The predominant experience of being unemployed is not one of long-term unemployment or work avoidance, but of 'churning' between low-paid, short-term employment and unemployment. People who have been unemployed for 12 months or longer in the last five years have three times the risk of in-work poverty.^[42] In 2013, just over 47% of unemployed people in Scotland had been unemployed for less than 6 months. UK-level analysis shows that almost half of men and a third of women making a new claim for Jobseekers Allowance (JSA) are doing so within six months of a previous claim.^[43] However, if benefit leavers are able to find relatively well-paid work, then their risk of experiencing a further period of unemployment is significantly reduced.

Low paid workers are less likely to progress

2.31 Low-wage work is a 'dead end' for many, with limited opportunities for progression. Low-paid workers are less likely to be offered training by their employer. Although some occupations and firms offer better advancement opportunities for low wage workers than others, workers who are low paid in one period are much more likely to be low paid later in life, regardless of other factors like skill level or gender. Over a quarter of UK low-paid workers in 2002 remained in persistent low pay over the whole decade to 2012 (a significant number of whom were cycling in and out of low-paid work), while just less than half moved out of low pay at some point only to move back into it, and only a fifth of low-paid workers moved into a consistently higher pay bracket.^[44] This suggests that the nature of the first job held by labour market (re)entrants is a key predictor of future earnings.

2.32 Many organisations and sectors lack clear career pathways or 'ladders', making it difficult for workers to understand where they sit in the labour market hierarchy, and what action they need to take to get to the next stage, thereby improving their pay and job quality. For example, focus group research conducted with workers in the early years sector found that higher qualifications did not seem to automatically equate to higher pay or status, since clear career ladders were absent in many settings.^[45] A number of individuals were keen to move into jobs within the wider children's workforce, such as teaching or social work, where pay and status was felt to be better. Their experience in an early years setting could have been usefully applied in these occupations, but they were unclear about the sectoral career pathways that could help direct them towards the necessary qualifications or experience.

Work intensity

Levels of part-time working and underemployment are high

2.33 Part time working rates vary between countries. Scotland has the third highest rate of part-time employment in the OECD (only Switzerland and the Netherlands have higher rates).

Chart 5 - Part time Employment as a Percentage of Total Employment (2012)

Source: OECD, Annual Population Survey

2.34 Between 2008 and 2013, the proportion of people in Scotland aged 16+ in full time jobs fell by 2.6 percentage points from 75.3% to 72.7%, while the proportion of people aged 16+ in part-time jobs increased by 2.6 percentage points from 24.7% to 27.3%.^[46]

2.35 The labour market has high levels of underemployment (people working fewer hours than they would like to) at present. The number of underemployed people has risen by 59,600 since 2008. This is particularly an issue for women working part-time.

Chart 6 - Underemployment by gender and working pattern

Source: Annual Population Survey, Jan-Dec, ONS

2.36 Underemployment levels are high in Scotland (and the UK) compared to other EU and OECD countries. Underemployment levels vary across Scottish local authority areas, tending to be related to unemployment levels, and therefore reflecting the degree of demand in the labour market.^[47]

Women are more likely to work part-time

2.37 Women are strongly over-represented in part-time work: 43.4% of employed women working part-time compared to 13.4% of men. Much (though not all) of the gap in income between men and women is due to women's shorter working hours, which in turn is partly a reflection of weak financial work incentives (determined by the interplay of the tax and benefits systems) and the lack of affordable childcare^[48], as well as choices families make to balance family and work.

Caring responsibilities are a barrier to increasing amount of paid work done by household members

2.38 For women in particular, their ability to increase their hours, or to enter/remain in work at all, is dependent on the availability of flexible working and affordable support for caring, for children or for someone with a long-term illness or disability.

2.39 Parents' employment decisions are influenced by the type of employment available, the financial benefits of moving into work, the costs and nature of childcare available, and perceptions about all these; and they need quality part-time and flexible work. Different factors may be relevant for mothers compared with fathers, and their constraints may also differ. For example, long hours worked by (usually) fathers, particularly prevalent in the UK, makes it hard for them to share family care, and this affects mothers' 'choices' with regard to the kind of work they can do.

2.40 There are an estimated 657,000 unpaid adult carers in Scotland, 62% of whom are women and over half of whom are of working age. Three in five people will become carers at some point in their lives.^[49] In the 2011 Census, 9% of people provided unpaid care for someone because of long-term physical/mental ill-health or disability or problems related to old age. Although this figure has shown little change since 2001, 44% of these unpaid care givers provided 20 or more hours a week of care a week, an increase of 7 percentage points since 2001.^[50]

Childcare is expensive

2.41 Pre-school care available out with the free entitlement is expensive, with Scottish parents facing some of the highest childcare costs in Europe.^[51] In the UK, 73% of mothers who do not work or who work part-time because of inadequate childcare services state that they do so because they consider childcare services too expensive, compared to 53% of mothers across Europe.^[52]

Availability and accessibility of childcare is just as important as the price

2.42 In 2013, there were significantly fewer childcare services in the most deprived areas compared to the least deprived areas.^[53] In 2014, 37% of Scottish local authorities (compared to 40% of English local authorities) indicated that parents had reported a lack of childcare in the preceding 12 months.^[54] The gaps in provision of childcare in Scotland were most marked for provision for disabled children, those living in rural areas, and those children whose parents have atypical work patterns. This is likely due to the higher costs of delivering these services and the inability of many families to meet this cost.

Welfare and tax

In-work poverty is costly in terms of welfare payments

2.43 Support for those on low pay is partly given through tax credits. Working Tax Credit (WTC) is available to low-income working households, while Child Tax Credit (CTC) is available to all low-income households with children whether or not they work. Their introduction has been identified as a major factor in the reduction of child poverty in the decade from 1998, with Scotland recording some of the greatest reductions in the UK. The situation has now changed. Living standards have fallen significantly since the financial crisis, and many are concerned that large parts of the tax credits system are effectively subsidising employers offering low pay.

2.44 In 2013-14 expenditure on WTC and CTC in Scotland was £2.2 billion^[55].

The interaction between earned income and benefit receipts on household income can be a significant barrier to increasing employment for families

2.45 Most benefit receipts for working families are dependent on the amount of earnings at the household level. The interaction between earnings and benefit receipts depends on the family type, and whether there are children or not. The effective marginal tax rate (EMTR) is the percentage of any increase in earnings that is lost through personal taxes or reductions in means-tested benefits. High EMTRs create weak incentives to move into employment and/or increase hours worked. High EMTRs mean increasing earned income may have little or no impact on total household income, leaving a household in poverty despite the increase in employment.

2.46 EMTRs vary markedly by income bracket and household type. Higher EMTRs are experienced by a higher proportion of households at the lower end of the income distribution than those at the top. At the lower end, lone parents and one earner married couples with children have very high EMTRs.

2.47 One strategy to reduce in-work poverty for single breadwinner families would be to support more second earners to work. However, the work incentives within the benefits system are not strong for this group, and the evidence suggests financial incentives are particularly important for second earners in low income households to move into work or work more hours.^[56] The Institute for Fiscal Studies (IFS) estimated the impact of Universal Credit on household income and poverty. They found that, overall, the new Universal Credit will provide stronger work incentives for more people than the current system. However, they estimate that it will weaken incentives to work for (potential) second earners in couples, because it has no earnings allowance for second earners, meaning the value of Universal Credit payments reduces as soon as the second earner enters work.^[57] They will see their returns from work fall from 59p to 35p per pound earned.^[58]

2.48 The interaction of the childcare system with work incentives means that the cost of entering work for second earners is much higher in the United Kingdom than the OECD average after childcare costs are taken into account. The UK's system of supporting childcare for working families is targeted at the low paid and lone families. For couple parents each earning 100% of the average wage in 2012, out of pocket childcare costs represented 29% of net family income. For a lone parent earning 67% of the average wage, out of pocket childcare costs were 8% of net family income.^[59] Reducing the amount of disposable income ordinary working families spend on childcare is an important part of lowering the barriers to work for women. Lower costs reduce the extent to which paying for childcare compounds the disincentives to earn than are already built into the tax and benefit system. Addressing the cost of childcare would also help to make work pay for lone parents.^[60] This is especially important where they are likely to enter low paid work: the cost of childcare can be the difference between work paying and paying to work for these families.

2.49 It has been proposed that 'better off in work' calculations should also consider 'time poverty'. By choice and/or necessity, many families are only 'partly working' to allow them to fulfil their parental duties (and enjoy the company of their children). For some, more time may be preferable to more money, especially if the work is not fulfilling and most of the extra income is spent on childcare.^[61]

The interaction between earned income and benefit receipts on household income can be a significant barrier to progression

2.50 High EMTRs are also a major disincentive to raising productivity through skills acquisition or other forms of investment. A worker is bound to be less inclined to go the trouble of acquiring an extra skill if 70% of the higher pay that might result is lost through personal taxes or reductions in means-tested benefits. Similarly, this disincentive impacts adversely on the employer who wants to share the benefits of a productivity-raising investment with their employees.^[62]