The living wage as an income range for decent work and life

Introduction

Most countries regulate pay through “minimum wages”, usually by law but also through collective bargaining (International Labour Organization, 2013). Wage minima may vary by sector, region or individual criteria such as age but the arrangements are usually designed both to provide a degree of income protection for workers, in combination with transferable benefits, and to prohibit “unfair” competition based on labour exploitation.

The declining wage share, rising inequality and increased cost of living observed across the developed world has recently shifted attention to “decent” income levels beyond minimum subsistence or market-clearing rates (Werner and Lim, 2016). This broader notion relates to quality-of-life and the ILO’s conception of “decent work”, which is also linked to work intensification and insecurity (Fleurbaey, 2015). In this context, a “living wage” (LW) is usually defined in terms of enabling meaningful participation in society beyond mere survival through, for example, recreation, supporting a family and saving against future risks (Clary, 2009; Ryan, 1996). Implicit in this notion of a LW is that not only is there an income level below which people risk further deprivation; above a certain threshold, there should be a qualitative upward shift in human freedom and capability (Parker et al., 2016). This broader perspective of a LW also recognises the impact of work-related factors on workers’ well-being (Waltman, 2004). Perceptions of workplace equity and justice, job satisfaction and life satisfaction generally can be regarded as indicators of human “thriving” that are theoretically enabled by a LW (Carr, Parker, Arrowsmith, Watters and Jones, 2016).

A fundamental problem of any form of standardised wage regulation concerns the diverse circumstances of the employment parties. Arguably, this complexity is more problematic in the case of LWs since they are usually calculated at a significantly higher rate than the usual minimum (Anker, 2011).

Empirical evidence of the real or hypothetical effects of a LW is limited and dominated by econometric analyses that focus on organisational and social outcomes. For example, recent research has examined the effects on poverty reduction (Neumark and Adams, 2003), economic development (Freeman, 2005), employment levels (Fairris and Reich, 2005), and social welfare policy (Chandler, 2009). Generally, LW studies tend to assume that employers are hesitant about LWs because raising pay will increase their costs and reduce investment (Karjanen, 2010). However, the implications for employees are less well understood, particularly along subjective dimensions such as perceived equity, job satisfaction and overall well-being. This means that one of the arguments for an LW, that higher pay can (as indicated by efficiency wage theory) increase skills development, retention and discretionary effort, remains under-developed and tested (Carr, Parker, Arrowsmith and Watters, 2016).

This paper addresses some of these issues through a qualitative analysis of employee perceptions relating to pay and aspects of work-life quality. It adopts a broader notion of a LW using a “capabilities” framework to explore individual needs, circumstances and preferences and provide empirical evidence concerning the potential implications of LWs as perceived by employees themselves. A literature review follows, focussing on our understanding of the LW concept and the utility of the capabilities approach. The next section outlines the study methodology, data collection and analysis procedures. Findings are then reported and used to inform a discussion of the potential impacts of a LW on individual and wider well-being. The paper concludes with a consideration of context and implications for future research.

Dimensions of a LW

A LW can be defined as dependent on context, shaped by objective considerations such as the cost of living and subjective (culturally or historically specific) expectations of needs. In practice, a LW is often defined and calculated on the basis of econometric analyses that use basic cost of living estimates, income distribution as a percentage of median income, or a combination of both (Carr, Parker, Arrowsmith and Watters, 2016). This notion of a LW rate is usually based on narrow assessments of the basic economic needs of a “typical” household unit (Anker and Anker’s (2017) recent definition is more workable but is still not broad). Though useful in itself, this neglects employee agency. A broader perspective would approach the LW in terms of employee understandings and impacts. In this paper, we define a LW as a wage level at which employees perceive and experience a step-change in their capability to enjoy meaningful organisational, personal and social lives.

Pay rates are not simply the outcome of “market forces”. Notions of what is fair and appropriate inform pay-setting structures and outcomes as much as labour market supply and demand due to the potential for conflict and need to elicit productive behaviour (Arrowsmith, 2009). Conflict may arise because pay is the means of livelihood for employees yet a foremost cost to employers; however, this is attenuated if pay correlates with productivity such that increases are offset by reductions in unit labour costs. Pay has potentially important behavioural effects because much effort and performance is exercised voluntarily (Colling and Terry, 2010), and perceptions of “decent” levels and processes of pay combine with job content, conditions and hours to shape employee commitment and work motivation (Stevenson and Wolfers, 2013). Pay is thus not simply a function of the external labour market but reflects the need of employers to recognise employee concerns in order to motivate discretionary effort (De Saá-Pérez and García-Falcón, 2002; Jawahar and Stone, 2011).

The LW concept incorporates but goes beyond such instrumental performance considerations by focussing on quality-of-life as a goal in itself. This perspective is grounded in traditions of morality and humanism (Bennett, 2014), reflecting an age-old idea that “wages should be sufficiently high to enable the labourer to live in a manner consistent with the dignity of a human being” (Ryan, 1906, p. vii). This dual focus on human well-being and productive development closely resembles the capabilities approach that has been applied across the social sciences (Deneulin, 2009). In Sen’s (1999, p. 75) seminal account, a person’s capabilities are essentially the meaningful freedoms he or she enjoys to lead the kind of life, and quality-of-life, they have reason to value.

Essentially, capability does not depend on the actual resource endowment that individuals may have (such as income) but rather on what people manage to achieve with their resource endowment given the freedom to choose through the exchange of resources. A human capability approach to LWs thus emphasises the impact of income on the autonomy of workers’ choices in the exchange of resources such as time and effort for money, as well as of money for goods and services. These choices, which are context specific, help to shape work motivation and perceived life quality, themselves subjective measures of individual capabilities.

Operationalising a LW using the capabilities approach is not straightforward given the diversity of individual characteristics and context and because there a large number of constructs and measures that can be used, whether for individuals (Waltman, 2004) or incorporating wider understandings of well-being such as at the household level (Stabile, 2008). These considerations of capability acknowledge not just material conditions but also sense of empowerment, commitment, psychological contract and job or career satisfaction (Clark, 2005). Ideally, an evaluation of the effectiveness of a LW in terms of its impact on human capabilities should involve broader considerations of a holistic evaluation of individual, family, organisational and social well-being.

We attempt to adopt this broader view here, accepting work motivation and perceived life quality as subjective indicators of individual capabilities which may be related to pay (Chiu and Chen, 2005; Riketta, 2008). Similarly, the organisational psychology literature highlights the relevance of factors such as wage relativities in informing perceived job satisfaction and well-being (Brown et al., 2008). Here, we use an exploratory capability approach based on selected indicators of work and life well-being. It is not our intention to utilise an exhaustive list of potential “capabilities” and “functionings” (in Sen’s (1999) terminology, the desired outcomes of a person’s capabilities), but rather to examine how income levels might relate to human capability or, more simply, perceived quality-of-life. This might be understood as qualitative since individuals define, for example, their sense of accomplishment or purpose in distinct family, work and community settings, in accordance with their goals (Clark, 2005).

A key concern regarding a LW, and an objective of this paper, is thus to explore the actual and relative levels of income that impact on workers’ perceived realisable opportunities, that is, their capabilities. More specifically, we explore the meaning of a LW based on qualitative insights as reported by employees in order to address two research questions drawn from the human capability perspective:

Methodology

The research was based on an online survey delivered in 2014 through various media platforms, details and overall quantitative analyses of which have been presented elsewhere (Carr, Parker, Arrowsmith, Watters and Jones, 2016). A sample of 1,183 employees participated in the survey under conditions of informed consent and confidentiality, following a low-risk screening process framed by University Ethics Committee protocol. Respondents were invited to respond to an open-ended question where they were asked to evaluate their wage or salary and provide views on the factors impacting on work-life quality. A total of 712 (60.2 per cent) participants provided comments but 106 of these were removed due to irrelevance or incompletion. This produced 606 usable participant narratives for this qualitative study.

The sample was diverse in terms of age, region, gender, occupation and income level though no claim is made for representativeness. For example, 41 per cent (n=249) were aged between 51-65 years (compared to 28 per cent under 40 years) and 24.4 per cent (n=148) were male; the former might reflect a tendency for younger respondents to reply via social media, which circumscribed narrative comment, and the latter a higher concentration of females in low-paid work (New Zealand Ministry for Women, 2016). Promotion of the survey was targeted at lower-waged workers (though community groups, trade unions and media appeals) but job status and occupation ranged from part-time, entry-level jobs to managerial and professional roles. Annual individual salary data thus ranged from New Zealand Dollar (NZD)3,000 to NZD400,000 (mean=60,787; SD=34,773). This range of pay and job types did, however, facilitate analyses of respondent perspectives concerning the different contextual relationships between pay, perceived work-life quality and job characteristics.

A template analysis of the data was adopted (King, 1998); this approach lies between content analysis, which applies a predetermined list of codes (Krippendorff, 2004), and a grounded theory approach in which themes surface within a loosely structured framework (Glaser and Strauss, 1967). The qualitative analysis software NVivo 11 was used to classify, sort and arrange the narrative data and to examine relationships between the themes that emerged. A two-step analytical process was adopted. The first stage involved “descriptive coding” (Richards, 2009). In order to understand the potential “pivot range” of income related to capability mobility, participants’ perceptions of the sufficiency of income was coded according to classificatory nodes extracted from the qualitative data. These nodes/categories were obtained using the search function of NVivo 11 to extract the most frequently used word(s) to describe subjective views of the quality of income. This approach generated six categories that we labelled, based on the content of responses, “struggle” (frequency=227), “barely enough” (109), “low pay” (203) “fair” (246), “comfortable” (99) and “really good” (84). These nodes were then used for coding. This is, of course, an interpretive rather than objective process, and the challenges of categorising narrative data into usefully descriptive and explanatory “variables” are familiar in qualitative research (Black, 2006), especially as it is important to “let the participants speak for themselves” (Maykut and Morehouse, 2005, p. 42). In order to ensure consistency during the analytic process, working definitions of each variable were generated and applied based on common respondent references to basic material and lifestyle factors, and research team members independently categorised and where necessary discussed each comment. Any coding discrepancies were reviewed and adjustments made until consensus was reached among the coders.

The second process involved “topic coding” (Richards, 2009) to develop understanding of the factors impacting on perceived work-life quality. A list of nodes was generated, in a similar fashion to the above coding process for salary references, concerning perceived capabilities and wider contextual factors. Coding was then applied to explore the components of these topics, based on participants’ comments. The coding framework was refined through a collective process of reading and re-reading the self-completed documents. Nodes were then compared and merged to form categories, with each category thoroughly analysed to identify recurrent patterns and themes. The “constant comparative method”, which is commonly used in grounded approaches (Lofland et al., 2006) was used to compare how different themes were discussed by different participants. This process served as an internal validation tool to enhance the credibility and potential transferability of the research (Guba and Lincoln, 1994).

Findings

The results are organised according to the two research questions. First, participants’ perceptions of income effectiveness are explored and, in particular, income ranges in which participants might move from “survival” to perceived “decent” income resulting in a transformation of the perceived capability. Second, contextual factors which impact on perceived work-life quality are explored. In both sections, quotations are used to illustrate important themes.

Discussion

Analysis of participant narratives reveals both unique and shared themes, pointing to links between pay and the dynamic capabilities that shape perceived work-life quality. Capabilities understood in this way become a basis for assessing equality of opportunity in terms of an individual’s ability to choose and achieve what he or she has reason to value. The narratives provide glimpses of self-perceived capabilities and ultimately what makes life experiences valuable (Sen, 2009). For example, many employees, especially those on low incomes, reveal the constraints on their personal choice not just in terms of wage expenditure but in wage generation, and not simply in financial terms but as being employed in jobs that do not satisfy their intrinsic or perceived future needs.

This paper uses the categorisation and reconciliation of qualitative data around pay and broader perceptions of capabilities to explore how employees connect pay to quality of (working) life. One finding was the suggestion of a pay range around which employee capabilities may be significantly changed within a specific (New Zealand) context. A pivot range of between NZD30,000-40,000 was indicated whereby people are more likely to perceive income in “decent” than “survival” terms. This shift from struggling to make ends meet to a relative well-being zone suggests a potential LW which enables individual capabilities. Interestingly, the findings were consistent with the 2014 LW hourly rate of NZD18.80, as campaigned for by the LW Movement Aotearoa New Zealand which is approximately NZD33,400 per annum for a full-time worker. The then statutory minimum wage was NZD14.25 per hour, or approximately NZD25,300 per annum.

The modal response category for each of the three highest income ranges was “fair” rather than comfortable or “really well” (Table I). This might emphasise the subjective nature of pay evaluation and the need to consider how individuals themselves interpret pay in terms of reward for and enabler of the application of capabilities. At the same time, for some individuals perceived needs and relativities might adjust to income so that real satisfaction is difficult to achieve. As one participant, who earns NZD167,000 explained:

It [my pay] doesn’t work for me well. I wanted to change my boat but wasn’t able to do it this year

(Male, 51-65 years, senior manager).

This paper offers an analysis of employees’ own perceived efficiency of income in terms of capabilities. The focus on capabilities as a measure of the personal and indeed social effectiveness of remuneration provides some practical insights into a potentially impactful LW pivot range, but is equally important in furthering understanding of the potential quality of a LW by exploring the social embeddedness and contextualised nature of work and wages. For example, many studies have given due attention to the importance of relative as well as absolute income in determining perceived work-life quality (Kifle, 2013). Our qualitative findings support this, highlighting that “fairness” is widely used to compare and evaluate wages – and jobs. This extends beyond pay to embrace a package of perceptions around treatment at work. Job security, for example, was a major concern to emerge as an important contextual factor in the research, linked as it is to income security but also perceived value at work. There are clear linkages between job security and work motivation (e.g. Sverke et al., 2002; Reisel et al., 2010). Our findings go further, suggesting specific, wage-related job security concerns that impact on individual perceived work-life quality. For example, some part-time workers were satisfied with their hourly rate but the uncertainty of working hours created significant stress. Others found the fragmentation of working schedules disruptive and incurring costs in transportation and dependent care. Our findings are in line with the argument that economic precarity such as job insecurity and poor work-life balance is a major challenge experienced by the working class (Warren, 2015). Pay rates cannot thus be considered in isolation from working time and working conditions. This leads us to call for a more holistic consideration of the potential impact of a LW.

Conclusion

This paper contributes to an understanding of the LW concept and practice by utilising an exploratory approach based on a capabilities theoretical framework and categorical analysis of employee narratives. In response to two research questions, we first examined and identified a pivot range for a LW, highlighting a discernible income range (NZD30-40,000) within which employees might perceive a step-change in work and personal empowerment. The findings complement business-case arguments that a LW can act as an “efficiency wage” (Marshall, 1920) and draw attention to spill-over and reinforcement effects between work and non-work.

These different contextual factors were explicitly addressed in response to question two. The findings indicated that respondents adjudged the effectiveness and legitimacy (“fairness”) of pay in multi-dimensional terms, relating to perceptions of personal inputs (skills and attributes), treatment at work, pay relative to others and according to their own contingent household needs. All of these factors provide the context for the impact of pay on well-being. However, it was also suggested that increased income, especially at and above the “pivot (LW) range”, has the capacity to deliver multiple gains outside the workplace as well as within in terms of enhanced capabilities and satisfaction. The findings tentatively suggest that a policy context framed by narrow economic cost-benefit analyses (e.g. wage costs vs employment) might miss the positive social externalities generated by higher pay.

Opportunities exist to extend our research. First, we acknowledge the generalisation problems associated with qualitative, exploratory studies, especially with relatively small (cell) numbers. More specifically, related to our methodology, we accept that the categories used are not to be reified and may be tendentious but they are generated and deployed in order to best capture and represent the overall subjective evaluation of pay efficiency by the participants themselves. The exploratory analysis of a body of narratives may be effectively supported by categorisation but deeper examination might be enhanced by alternative approaches such as semi-structured interviews.

Second, Figure 1 is provided for illustrative purposes to support the empirical finding that, notwithstanding differences in personal circumstances, a “pivot range” might be identified that distinguishes basic from decent income in terms of potential thriving. The approach was to apply a thematic analysis via manual and NVivo techniques to identify any consistencies within the diverse sample that shed light on the link between income and self-perceived capabilities. No claims can therefore be made about statistical rigour. Furthermore, this research was conducted in New Zealand. Although its insights will be most relevant to similar country contexts, the focus on the relationship between income and capabilities is of more general concern. Subsequent comparative studies could provide more knowledge on the context-specificity of a LW.

We hope our findings offers insights into the dynamic nature of the impact of incomes on individual perceived capability, and future research might apply different methods (e.g. surveys, mix-methods) to test and extend our results. Ideally, too, information about income and career transitions over time would be beneficial. In particular, longitudinal research would enhance understanding of whether a LW has short-term or enduring effects on individual, family, workplace and societal well-being, and could fruitfully inform policy debates around pay.