Abstract
The ‘living wage’ is an idea with a long history in the UK currently enjoying a renaissance. This article explores possible reasons for its re-emergence as a policy demand, but argues that thinking of low pay primarily as ‘poverty pay’ caused by employers’ failure to pay a living wage raises practical and conceptual issues that are problematic. It examines to what extent recent attempts to resolve such issues in the UK and elsewhere have succeeded, and concludes by suggesting that alternative ways of analysing and addressing the two key issues associated with the living wage, low pay and in work poverty, are required.
Africans want to be paid a living wage.
… the theory of a living wage based on an average family is the greatest statistical fallacy of this or of any other age.
Introduction 1
This article discusses the ‘living wage’ – an idea with a long history in the UK (Snowden, 1912; Grover, 2009) and elsewhere (Stapleford, 2008) currently enjoying a renaissance. Ryan (1906: vii) described it as the idea that ‘wages should be sufficiently high to enable the labourer to live in a manner consistent with the dignity of a human being’. Ada Nield Chew, a working class mother and activist writing in 1894 about factory women, put it graphically: ‘We cannot be said to “live” – we merely exist … A living wage! Ours is a lingering dying wage’. 2 She conveys clearly the appeal of calls for a living wage, and also why it is typically seen as including a margin above the ‘poverty line’, to ensure a socially acceptable and decent lifestyle.
Ryan also discussed whether a living wage should be calculated only for the labourer, or should also include the needs of any family s/he might have. Campaigns often call for not just a higher hourly wage but also improvements in paid holiday and sick leave, and union representation (Pattison, 2009). ‘Living’ here clearly goes beyond just income. This is a different approach from that of the Low Pay Commission (LPC, 2010), which in examining the UK’s national minimum wage (NMW) level is required to have regard to the economy and competitiveness, but not to ‘what is actually required to live’ (Wills, 2009: 38).
Writing on the living wage often largely discusses higher pay for low-paid workers, 3 rather than the living wage itself (e.g. Wills and Linneker, 2012). But this article does not discuss general arguments about the possible negative impact of higher wages (on affordability, competitiveness, unemployment etc.), or claims for their positive effects (increases in living standards, self-esteem and work–life balance for employees, or improvements in productivity, recruitment and retention for employers), as these apply to any strategies to improve low pay. 4 Neither does it chronicle the development and successes of the living wage campaign, which are described elsewhere (e.g. Living Wage Foundation, 2012). Instead, since the living wage is ‘under-interrogated’ (Resolution Foundation, 2012), this article explores not only possible reasons for its re-emergence but also the issues raised by conceptualizing low pay primarily as ‘poverty pay’. (One Greater London Authority (GLA) press release, for example, claimed the campaign had ‘lifted 45,000 people out of working poverty’ (GLA, 2012b).) It examines to what extent recent developments have resolved these issues. And it suggests that alternative ways of thinking about low pay and in work poverty would be less confused conceptually in terms of analysis, and more coherent practically as a basis for policies and strategies.
It was in 2001 that the reincarnated living wage campaign started in London, led by part of the organization London Citizens, and championed by the GLA. The Fair Pay Network helped to spread the word across the UK (Littman et al., 2010), and the Living Wage Foundation now accredits living wage employers. A recent report argued that big firms in various sectors can afford to pay the living wage (Pennycook, 2012); and the UK’s (now) prime minister described it in 2010 as an idea whose time had come.
In such a climate, and given the longevity of the idea (Anker, 2011), it is difficult to ask whether the living wage is either an intellectually coherent concept or the best focus for action on low pay. But several issues arise if low pay is conceptualized primarily as employers’ failure to pay a living wage. These include framing low pay in terms of individual/family poverty, rather than unequal rewards in the labour market; and (in some formulations) the assumption that, if one worker in a family earns enough, the dependence of another adult on them is not problematic. More technically, the number of assumptions needed to translate an hourly gross pay rate into a weekly disposable income for an individual/family, to arrive at a living wage figure, means that the resulting amount is unlikely to match all workers’ circumstances. In addition, calls for a ‘living wage’ may imply that it is primarily employers’ responsibility to meet the needs of workers and their families; but ‘if creating a living wage is seen as primarily or largely a government responsibility, then a wider array of policy tools are available’ (Weldon and Targ, 2004: 90).
Some living wage supporters call (as in a 2007 petition on the 10 Downing Street website) for the NMW to be raised to a living wage level (see Waltman, 2004, for the US). Others instead ask employers – including government departments, but also private sector companies – to pay the living wage voluntarily and ensure that their subcontractors do so. Similarly, a clause in the Ethical Trading Initiative’s (ETI’s) ‘base code’ requires companies in global supply chains to pay wages sufficient to meet basic needs, with a margin for discretionary income (Ethical Trading Initiative, 2012). However, the living wage is also used more loosely. Many probably understand it as just a slogan; but its supporters claim it is more effective than demands for ‘fair pay’ or a ‘decent wage’.
Different living wage amounts may be demanded. But one difference is an issue of principle. Some campaigners believe that (full-time) workers should be able to earn sufficient to support themselves and their families without recourse to means-tested in-work benefits/tax credits (e.g. Fair Pay Network, 2012). Others, however, assume that any in-work benefits/tax credits are claimed. Grover (2008) argues that the demand is therefore in practice for a living income, not a living wage. These approaches result in different figures; the GLA (2012a) cites both. But they always argue for higher pay on the basis that workers (and their families) cannot reach a certain income level on their current wages; they therefore see low pay as ‘pay that is too low to allow a worker and their dependants to be free of poverty [and some would add] without means-tested support’ (Howarth and Kenway, 2004: 2).
Reasons for renaissance?
Why, a century or so after the living wage last figured seriously in UK policy debates, has it recently re-emerged? Wills (2009: 33) links this to the renewed ‘sweating’ of labour (Standing, 2011), and the economic crisis as a catalyst for new ideas. The growing focus on the ‘activation’ of unemployed and, increasingly, ‘inactive’, groups (OECD, 1994) also leads logically to concern that work should pay for those affected (Grover, 2009). In addition, the UK’s Fair Wages Resolution, governing terms and conditions for contract workers for public bodies, was abolished in the 1980s, since when contracting out by public authorities has significantly increased. And, as trades unions have declined, their difficulties in organizing migrants and others in ‘vulnerable employment’ (CoVE, 2008) left the door open to alternatives (Holgate, 2009), with the vibrant, consensual, charismatic and conflictual organizing style of Citizens UK (Rutherford, 2010) offering an antidote to party politics (Baskerville and Stears, 2010). The more fragmented nature of the labour market may suit such localized activity, which can ‘name and shame’ employers more effectively, and which is often carried out jointly with unions (SERTUC, 2007).
UK activists may also have been inspired by campaigns in the US for higher pay for employees of companies subcontracted by public authorities (or in receipt of public funding, or on government land). Such campaigns often mobilize local communities (Freeman, 2005), and Citizens UK has close relationships with some US churches involved. Internationally, ethical consumerism is growing in importance, and a living wage for workers in the global south is seen as an extension of the fair trade model, and a way for brand companies to avoid reputational risk (Miller and Williams, 2009).
The living wage calculation
However, to campaign for a living wage it is necessary to know what this means in practice. A series of steps is needed to move from the weekly disposable income needed by an individual/family/household to raise them above a certain threshold to the gross hourly pay for the individual worker(s) required to reach that income. Hirsch, who calculates a living wage level outside London, recognizes that this is ‘far from straightforward’ (2010b: 2). Key issues include what kind of household the living wage should support; how many earners working how many hours should be assumed; and what should be done about certain variable costs.
Household unit, number of workers and hours of work
Using different sizes of family, with different numbers of earners working varying hours, in the living wage calculation produces different figures. The approach comes closest to a ‘model families’ method, using one model family or a range.
In 1918, Seebohm Rowntree called for a living wage sufficient to support a one-earner couple and three children. Eleanor Rathbone (1924), pointing out that this would still leave larger families unable to meet their needs, whilst providing for millions of ‘phantom children’, argued for family allowances (now child benefit) to be paid outside the wages system. Some supporters use a ‘typical’/‘average’ family of a couple and two children – also used in the US for some time (ACORN, 2003). Labour Behind the Label identified a ‘floor [living] wage’ for Asia by calculating the cost of providing a daily number of calories for two adults and two children, adding non-food costs, and averaging the amounts over different countries (Luce, 2009). 5 Some US living wage calculators only assume one (full-time) worker, but attach him/her to various different households; the reference household for some, however, seems to be one adult and one child. 6 Church Action on Poverty used to use a full-time single worker. In 2001, TELCO used a couple with two young children, with one full-time and one part-time earner. (The GLA also used to calculate a lower young person’s living wage, but no longer does so.)
But in some formulations, the living wage is a traditional ‘family wage’ (Land, 1980), based on a couple – usually with children – with one full-time earner. Feminists argued against this in the 1980s (TURU, 1981; Family Wage Group, 1982) because it was based on a ‘male breadwinner’ model, implying that, if the main earner’s wage was sufficient to maintain their family, it was acceptable for the other adult to be out of the labour market, or paid ‘pin money’. Grover (2005) argues that the idea of need in connection with the living wage is problematic because of this association. The approach was also vulnerable to attack by employers, who only had to show that their workers were typically not male breadwinners but part-time married women to undermine it. So basing wage demands on family responsibilities failed to help these low-paid groups to raise their pay. Figart et al. (2002) argue that it is now widely accepted that women support dependants too, so the demand is for a gender-neutral family-sustaining wage. But this still does not deal adequately with economic dependence within the family.
To overcome the problems of using one standard family, the GLA takes a variety of family types (with different assumptions about number of earners), and weights the results according to the proportions of such families in the London population. It assumes that full-timers work 38.5 hours/week and part-timers 17 hours (GLA, 2012a). This method has recently been modified by the Centre for Research in Social Policy (CRSP) (discussed below). The problem is not resolved, however, by averaging over a selection of family types, because a living wage based on average needs always means some will find it insufficient. The GLA, for example, excludes families with more than two children, or older children. The Living Wage Foundation (2012) website claims that ‘a living wage means every worker in the country will earn enough to provide their family with the essentials of life’. But this is clearly true only in certain cases – which, as Grover (2008: 77) argues, is ‘at odds with the idea that the GLA’s approach is responsive to need’.
Variable costs
Other elements of the calculation include assumptions about housing costs and local taxation, as well as what, if any, childcare costs are incurred etc. These costs are likely to vary in different areas (and, for the global campaign, in different countries).
Impact of government policies
A wage may not be the only income entering a household. But living wage campaigners in the UK argue that many low-paid workers are migrants, who may have limited or no access to benefits. And in the global south, families must often pay for education and/or health care. It is perhaps unsurprising that strong living wage campaigns are pursued in the US, where public expectations of government action tend to be much lower than in Europe – indeed, a prime motivation for some living wage campaigns seems to be to ensure that workers and their families can live independently of state support (Wider Opportunities for Women, 2012).
But regardless of ideological stance, the living wage level is inevitably affected by government policies. Annual wage rises are traditionally negotiated in part in relation to inflation, which is influenced by government. But the living wage is more susceptible to a wider range of policy changes. If direct taxes are reduced, and/or benefits/tax credits increased, the living wage level will be lowered. 7 If wages are seen as being about subsistence, rather than relative labour market rewards, this may make sense. But it has potentially problematic effects. It could be seen as making it harder to achieve higher net incomes for those at the bottom, since improvements due to government action will be counter-balanced by a lower living wage. And it could be more difficult to campaign for increases in the ‘social wage’ (benefits and services outside the pay-packet), or reductions in childcare or transport costs etc. – often particularly important for women – because the answer to inadequate income in relation to needs seems to rest primarily with employers. This is paradoxical since, as Weldon and Targ (2004) note, ‘wage supports’ such as childcare make living wage demands more attainable. And sustaining a long-term, coherent strategy on low pay could be harder, because the living wage figure is vulnerable to tax/benefits changes, rather than being seen consistently in relation to the whole structure of labour market rewards.
Problems resolved?
Some problems explored above have been addressed recently in initiatives in the UK and elsewhere, which are examined below.
UK: Refining the living wage calculation
In the UK, the focus has been on improving the living wage calculation. From 2010, CRSP put the methodology for calculating a living wage figure for areas outside London on a more secure footing, based on research into a minimum income standard (MIS) for the Joseph Rowntree Foundation (Hirsch, 2010a, 2010b; Hirsch and Moore, 2011). Several modifications were made to approximate more closely the GLA’s approach, though the longer-term aim is to have a common method.
In relation to the issues above, Hirsch argued that: while there is no single answer to any of these questions, this does not prevent us from seeking to identify a central figure for a living wage that comes as close as possible to meeting the minimum needs of most workers. (2010b: 3)
But, as he notes, the nine family types used, whilst more extensive than the GLA’s, are still a sample of cases, with certain assumptions about family size (single people, couples and families with one or two children), 8 working patterns (all adults working 37.5 hours/week) and other factors (e.g. ages of children, childcare costs, families with children accessing social housing etc.).
The assumption that all adults work full time means the living wage figure is not based on adult dependence, avoiding ‘family wage’ implications. The converse is pointed out by Hirsch (2010b: 6), however – many families with young children ‘do not adopt’ this pattern, and (he argues) ‘would not want to accept’ it. Neither does it seem to resolve the issue that initially motivated London living wage campaigners, which was parents’ lack of time with their children because of working long hours in low-paid jobs (Baskerville and Stears, 2010).
But in addition, the continued use of a ‘model families’ method means that we do not know how many families of this type, with these costs, and both adults working full time, exist and are living in poverty. We can guess, however, that the number may not be significant, as this does not accord well with the profile of low-paid people. And if we are concerned about workers in the flexible labour market and/or at the bottom of the wage distribution, full-time adult workers with children are probably not typical of these. In the US, ‘some ordinances do now provide a wage adequate to support a family with two children if both parents work full time. But this still leaves many families … without enough earned income to meet their basic needs’ (Weldon and Targ, 2004: 88); they conclude: ‘In picking an inadequate threshold, the movement gives up much of its rhetorical strength. The argument is that all workers should be able to meet their family’s basic needs, through work’ (emphasis in original).
The dependence of the calculation on experts might be seen as disadvantageous. In addition, living wage rates suffer from ‘volatility’, because the method ‘binds [them] to fluctuations in the prices of basic goods’, highlighting ‘the difference in behaviour of living wage … as opposed to relative low pay thresholds’ (Pennycook and Whittaker, 2012: 16). The campaign therefore agreed to limit both the rise in net income in relation to the increase for someone on average earnings and the increase in the living wage in relation to the increase in average gross earnings, whatever the figure produced by the formula (Hirsch, 2012). These limits were applied to the 2012 living wage figure outside London, with the £8.80/hour derived from the calculation labelled an aspirational ‘reference level’, and the level recommended instead being £7.45 (£8.55 in London). 9 This is a significant shortfall. There would be realignment over the longer term, and this commitment will be kept under review. But for now, it is hard to see the new methodology having emerged intact.
It is also unclear how employers will build the living wage into wage setting processes (though the figure will now be announced annually in autumn, to fit with spring negotiations). More importantly, Hirsch (2010a) has argued: while it makes a nice slogan, a ‘living wage’ can only thrive in the long term if people are convinced that it’s what it says on the box – that is, what people genuinely need as a minimum to live on.
The limits placed on the 2012 living wage figure clearly, however, demonstrate that other, pragmatic and political, considerations have taken precedence over this (understandably, it may well be argued).
What about geographical variation? The MIS calculation could include local variations in housing costs and council tax; but it excludes differences in other local costs. A key argument for the NMW was its simplicity, with one national rate assisting in enforcement. However, the current living wage campaign started with the argument that London had higher costs but the NMW had no ‘London weighting’; and taken to its logical conclusion, there could be different living wage rates in each region, town or even village. But UK campaigners believe the different ways of calculating a living wage in the US may have weakened the movement there, and so have agreed that locally calculated living wage figures must be approved by the Living Wage Foundation national board. This central oversight is somewhat at odds with the campaign’s emphasis on community involvement, however.
It is clear that the living wage is not, and cannot be, ‘what it says on the box – that is, what people genuinely need as a minimum to live on’ (Hirsch, 2010a), or an answer to in work poverty for all workers. As Weldon and Targ (2004: 90) conclude: ‘The problem is the conceptual basis for living wage movements: the idea that one wage amount can be specified that will constitute a living wage for all workers and their families’; they suggest reconceptualizing it as a package of wage and non-wage policies from employers and government. 10
Globally: De-emphasizing the calculation
Living wage campaigners could argue that, whilst there will never be a perfect figure, they need a wage level they can broadly defend; employers do not query it; and the key imperative is to improve wages for all. But as Miller and Williams (2009) argue, ‘debate continues to rage about living wage formulae’ (p. 101), with ‘no consensus on … a methodology’ (p. 105). The problem of averaging family needs was one reason why the ETI decided that local negotiation of the level by workers affected was preferable to a formulaic approach (Steele, 2000). It also argued that ideas of ‘the family’ vary in developing countries – and with additional resources, the number of people in ‘the family’ may expand. And it sees the London living wage calculation as ‘top-down’.
Local negotiation could be seen as delegating the resolution of problems – or, as an alternative solution, de-emphasizing the calculation, rather than introducing a more sophisticated method, as in the UK. The ETI website states: ‘It’s important that companies don’t allow the challenge of how to calculate a living wage to distract them from getting on with raising wages’ (Ethical Trading Initiative, 2012). The International Labour Office (ILO) adopted a ‘fair wage approach’, portraying the living wage as only one of several campaigning objectives relating to low pay (including the minimum wage and prevailing sector wage). The Fair Wear Foundation (2012) also developed a ‘wage ladder’ with similar aims, arguing that ‘progress has stalled in discussions about what, exactly, constitutes a living wage’.
Various ways round this are being pursued. The latest initiative returns to the calculation, in a review by Anker (2011) for the ILO aiming to work towards a consensus on definition and measurement. Anker does not question the concept of the living wage, but cites Ryan (1906) on the difficulty of being precise about a figure, and suggests the idea is ‘subjective’ (especially in relation to what level of costs should be covered), with measures therefore proving problematic. He demonstrates the lack of agreement on family size, number of workers etc. – though he says the living wage is a ‘family wage’, because even single workers may support family members elsewhere. He suggests assuming a household of four (curiously, because this represents the demographic replacement rate), and 1.5 full-time workers (who may be unable to work all year) – and in general, that technical calculations should be separated from political choices about the acceptable wage level and characteristics used to set it. Although there is no methodological silver bullet, he argues, improved measurement is a necessary condition for acceptance of the living wage. The implications are that, although the living wage campaign in developing countries had tried to de-emphasize the calculation, this is becoming too serious a roadblock to ignore. But it is not clear that Anker’s review will resolve the issues. And his identification of the ‘subjective’ nature of the calculation as the main problem is surely mistaken, as discussed below.
Low pay and in work poverty
The major reason for difficulties with the calculation is instead that conceptualizing the issue primarily as employers failing to pay a ‘living wage’ confuses low pay and in work poverty and the reasons for each (Bennett, 2012).
Low pay is usually measured in terms of individuals’ gross hourly pay (or possibly in relation to annual earnings for full-year workers – Marx and Nolan, 2012). It is often linked to other forms of employment disadvantage, and limited to employees rather than including self-employed people. Several journalistic exposes of low-paid work have been published recently in the UK and US (Toynbee, 2003; Ehrenreich, 2001). But who are the low paid? A recent report on Londoners paid under the living wage found that the main groups affected included young people and part-time workers (Smith, 2012); and a national analysis showed 71 per cent of 16–21 year olds earning below it (Savage, 2011). Pennycook and Whittaker’s recent report (2012) found one in five workers earning below the living wage nationally. (They also use a low pay threshold of two-thirds of gross median hourly pay, which largely gives similar results but is national, whereas the living wage is higher for London.) Workers who are female, part time, on temporary or casual contracts and working in the private sector are at greater risk of low-paid work; and young workers (aged 16–20) are also at high risk.
These are not the types of workers often envisaged in references to ‘poverty pay’, however. Describing pay as lower than the living wage evokes in work poverty, used to describe a family/household in poverty with one or more earners. For example, a recent report finds 6.1 million people in ‘working households’ in poverty in the UK (Aldridge et al., 2012). In practice, however, this does not always mean that the worker involved has a low hourly wage. Neither do all such low-paid workers live in households in poverty.
Here we look briefly at the complex relationship between low pay and in work poverty. There is certainly some overlap. Millar and Gardiner (2004) concluded that the overlap grew significantly in the UK between the 1970s/80s and 2000/01, from 3–4 to 14 per cent of employees. Cooke and Lawton (2008) argued that it had decreased again by 2004/05, to some 7.2 per cent, perhaps because of policy changes.
11
But: There is not a straightforward relationship between earning a low wage and living in poverty … because different family types need different amounts of money to enjoy a similar standard of living and because an individual’s own wages are only one source of income on which a household can draw to escape poverty. (Cooke and Lawton, 2008: 42)
However, it could be argued that like is not being compared with like. Both these studies used the household as the unit of analysis, whereas the living wage calculations use individuals and nuclear families; a household may be wider (e.g. young adults living with their parents). And, as the authors acknowledge, their studies imply full sharing of resources between members of the wider household, which may be unlikely.
Investigation of NMW recipients by Brewer et al. (2009) focuses on the family instead. Their report found that NMW workers tended to work part time and to be female, under 25, less educated, disabled, minority ethnic (especially Bangladeshi/Pakistani), and living in social housing and/or in the north of England, Wales or Northern Ireland. Families containing NMW workers, however, were unlikely to be at the bottom of the working age income distribution, and more likely to be towards the middle. (See Dickens, 1999; Sutherland, 2001; Manning and Dickens, 2002.) The three-fifths of families with an NMW earner for which this was their main source of earned income were towards the bottom (not the very bottom) of the distribution. But ‘families’ here includes individuals, who may be (e.g.) young people living with parents; the majority were indeed single childless people, though lone parents had the highest risk.
However, these analyses also all use income before housing costs. Poverty is often measured in this way, especially cross-nationally; but housing costs may be one reason for falling into poverty, and arguably should therefore also be examined. For the MIS living wage calculation, childcare costs are also critical for those with children – but excluded in these studies. In addition, full take-up of means-tested benefits and tax credits may be assumed. So we do not have comprehensive data from these studies about the overlap between low pay and family/household poverty directly comparable with the ‘model families’ approach of the living wage calculation.
In any case, a snapshot of household/family income may not be the most appropriate way to conceive of the key aspects of the relationship between low pay and poverty. Low pay is arguably not most important for someone because they happen to live in a poor family/household, but because of the higher risk of insecurity and poverty over their lifetime. Low pay may mean that they are not earning rights to future benefits, especially pensions; and they may be at greater risk of poverty if their relationship breaks down and they have no individual assets or savings. This is more likely for women than men; and it may be true not just for low-paid people living in households in poverty but also for others.
Rethinking low pay and in work poverty
The argument here is that neither a precision formula for a living wage nor de-emphasizing calculation can solve the conceptual conundrums caused by the confusion between low pay and in work poverty. Instead, there is a need to rethink how to approach both these real social problems.
Figart et al. (2002) argue that wages can be conceived of as a living, and as a contribution to production, but also as translating social values and prejudices into differential rewards (e.g. reflecting class, gender and ‘race’). An alternative way to tackle low pay is therefore to challenge these unequal labour market rewards, and to argue for a ‘fair wage’ or ‘decent pay’ for all. There is some support for this approach from international bodies. The Community Charter of the Fundamental Social Rights of Workers (1989) refers to ‘a decent wage’, and says that all employment should be ‘fairly remunerated’ (Clause 8). The ILO’s agenda to achieve ‘decent work’ for all began in 1999. And the Council of Europe’s ‘decency threshold’ is based on a percentage of median earnings. These are based on relative rewards in the labour market, focusing on inequality of wages rather than household poverty; and campaigners have argued that everyone should be paid at least two-thirds of the median wage. Sometimes the median male wage and/or median of full-time wages is used, to circumvent the traditional under-valuation of women’s work (e.g. European Foundation for the Improvement of Living and Working Conditions, 2010).
Several organizations have started to campaign on low pay in the context of inequality rather than ‘poverty pay’. The Equality Trust, founded following The Spirit Level (Wilkinson and Pickett, 2009), calls for higher minimum wages without mentioning the living wage, and campaigns for local authorities to narrow pay inequalities (‘Act Local Reduce the Gap’). One Society, a spinoff, included the living wage in a recent report, but largely focuses on inequalities in labour market rewards. And MyFairLondon, also linked to the Equality Trust, calls on the mayor to reduce the gap between the highest and lowest earners in the GLA group to 10:1, use procurement powers to influence suppliers, and lobby other London employers to make their pay more equal. These bodies outline a clear alternative framework for considering low pay. To date, however, their public profile has been low.
Krugman (2002, cited in Perrons, 2010) argues that social norms are now more tolerant of greater inequality; and that this, not just the onward march of globalization, trade and/or technological change, may be leading to increasing inequality of rewards. If so, Perrons argues, this allows for questioning current earnings distributions and examining the link between pay and the social value of different work – an upending of Figart et al.’s argument that wages reflect social prejudices; and this involves challenging the under-valuation of care jobs, usually done by women. The dignity that living wage campaigners demand for all workers arguably relates more clearly to the valuing of the work one does than to a living standard for the household in which one lives. Concern about low-paid workers would then focus on other workers than those in the ‘model families’ in living wage calculations, especially part-timers and young people.
In work poverty also requires rethinking. This is a portmanteau term, ripe for the forensic investigation that has now begun. There is some recent UK research (DWP, 2008; Kenway, 2008; Ray et al., 2010). But it is in European Union (EU) documents that the unpacking is largely being conducted, following the emergence of in work poverty as a key policy issue (e.g. Eurostat, 2010; 12 Ponthieux, 2010; Frazer and Marlier, 2010; Marx and Nolan, 2012).
The idea of in work (household) poverty is inherently unclear. As the European Foundation for the Improvement of Living and Working Conditions (2010: 2) put it: The ‘working poor’ … concept combines two levels of analysis: the working status of individuals (individual level) and the income status of the household in which they live, which is below the poverty threshold (collective level).
To identify the ‘working poor’ only as individuals neglects intra-household interactions that may shape labour supply behaviour (Ponthieux, 2010); but households are not ‘in work’, because they are not labour market entities.
The European Commission now includes a breakdown of its ‘at risk of poverty’ indicator measuring ‘work intensity’ (Bardone and Guio, 2005); this conveys how much work is being done in a household below the poverty threshold of 60 per cent of median equivalized disposable household income with one or more earners, and so starts investigating ‘in work poverty’ more carefully. 13 Further refinements include greater differentiation of family types and recognition of part-time work (Tarki Social Research Institute, 2010). The poverty threshold has risen because of increasing numbers of dual-earner households (Marx et al., 2012), meaning that one-earner households now run a high risk of ‘in work poverty’, and even low-paid work helps prevent poverty in dual-earner households. Using EU-SILC data, Marx and Nolan (2012) find in work poverty associated not with low hourly pay but with less than full-time or permanent labour market engagement at the individual level, and with single earners and low work intensity at the household level; but policy solutions will be very different depending on whether this problem is seen as either insufficient breadwinner earnings or partner non-participation in work.
Ponthieux (2010) takes this analysis further. She notes the difficulty in interpreting in work poverty because the same poverty rate may result from labour market and/or household factors as well as social and fiscal policies. In a study of 12 EU countries, she examines the relative influence of these on the risk of in work poverty. 14 The first stage is to develop an intermediary indicator of ‘poverty in earned income’, to examine individuals’ labour market position in relation to an individual poverty threshold. The second stage is to put the individual back into the household, and examine any transfers they receive. This two-step approach begins to disentangle some issues involved in analysing in work poverty. In particular, it demonstrates the crucial influence of gender, in terms of the significance of women’s employment situations (often hidden by household level analysis). In the bulk of countries examined, women’s in work poverty risk is related more to their own employment characteristics and low earnings, whereas men’s is related more to household factors (including their partner’s employment situation). This means that in work poverty is associated with institutional factors relating to decommodification and defamilialization affecting labour market participation (Marx and Nolan, 2012, citing Lohmann and Marx, 2008). The study by Ponthieux, although dogged by data deficiencies, and not examining all the costs included in a living wage calculation, begins to develop the tools needed to unpack ‘in work poverty’.
In this study, ‘earned income’ includes replacement incomes linked to temporary absence from the labour market. This therefore raises the issue of income available to a partner not in the labour market as relevant to combating in work poverty. But in the modelling approach often used to examine in work poverty, the workless adult in couples with one earner never seems to have any income of their own. It is clear, however, that if someone is out of work for certain reasons, they may – and arguably should – have their own income, such as a non-means-tested earnings replacement benefit (in the UK, for example, carer’s allowance, employment and support allowance, jobseeker’s allowance, or pay during maternity or parental leave). These can be crucial protective factors against the risk of in work poverty for individuals in couples (Bennett and Sutherland, 2011; Marx and Nolan, 2012: 12) – but are hardly ever considered in analyses of in work poverty or associated policy recommendations.
Interestingly, Ponthieux (2010) found that in the UK (and Sweden) over three-quarters of workers at risk of in work poverty were poor in earned income; and Marx and Nolan (2012) found that for single people, net income at minimum wage level was under the poverty threshold only in the UK (and Spain). Both findings suggest that – whilst still distinct phenomena – low pay and in work poverty may overlap more in the UK than in many other EU countries, and this may be one reason why the living wage is a live concept in the UK more than elsewhere in Europe.
Conclusions
Campaigners for the living wage see it as an idea which resonates widely with the public and employers and which should be scaled up. Some people believe instead that its main function could be more limited – trying to guarantee that contractors in the supply chain pay workers at least a minimum amount (e.g. Coats, 2007). And Metcalf (2007: 50) concludes that ‘essentially … the living wage is best viewed as a rallying cry to boost the pay of those towards the bottom of the wage league table’. There is evidence that in the UK recently the living wage has enjoyed some success in performing this function. And its supporters might argue that fairness is also a contested idea – and that if strategies to tackle low pay are based on tackling inequalities in labour market rewards, it will be harder to mobilize supporters or challenge employers; the idea of ‘poverty pay’ is more emotive, and has traction.
This article suggests instead, however, that seeing low pay through the lens of household poverty – defined as living below a relative poverty threshold, or more generously – can only ever provide a partial solution, because it does not place inadequate wages in the context of the unequal structure of labour market rewards and the persistent under-valuation of certain forms of paid (and unpaid) work. And in the longer term, this perspective is likely to be challenged by employers and others arguing that an individual’s hourly gross pay level has only an indirect relationship with the weekly disposable income of their household – as already happens in relation to the NMW. In addition, instead of employers being called upon to help all workers with their needs, averaged across a variety of different situations, via a ‘living wage’, they could be asked to support those workers who do have family responsibilities at the relevant times in their lives. Such needs are better met through specific employer provision – such as paid maternity/paternity/parental leave – and/or government policies such as social protection and affordable housing. The current welcome unpacking of the nature and causes of in work poverty also needs to be broadened, to foreground the key issues of the gendered division of labour and the presence/absence of social protection for adults out of work in couples.
