What Counts when it Comes to Salary? A Qualitative Analysis of the Arguments Dutch Citizens Use when Distributing Salary

Abstract

This study examines to what extent job evaluation systems align with public views on fair pay. Using a Participatory Value Evaluation, over 3,700 Dutch citizens distributed extra salary among various jobs and explained their choices. The arguments were compared to criteria used in job evaluation systems. The dimensions prioritized by participants, such as required education, job complexity, and working conditions, are already well represented in current formal job evaluation systems. However, participants also stressed aspects often absent from such systems, including a job’s societal contribution and labour market scarcity. While job evaluation systems remain broadly consistent with public reasoning, we recommend initiating broader public and institutional discussions on the evolving values that underpin fair compensation.

Keywords

empirical research fair pay formal pay systems salary public opinion pay practice job evaluation systems

Which job deserves what salary, and why? While formal job evaluation systems are designed to regulate this question, their outcomes ultimately depend on the criteria they apply. To ensure these systems produce fair and legitimate results, it is essential to assess whether their underlying criteria remain comprehensive, valid, and relevant. Understanding public values concerning fair pay can help ensure that the principles embedded in job evaluation systems reflect contemporary notions of fairness and legitimacy. Fairness is a key principle in compensation for work: perceived fair pay supports employee satisfaction and retention and contributes to an organization’s reputation and performance (Belcher, 1979; Scott, 2018). Conversely, perceived unfairness in pay can undermine morale, reduce productivity, weaken cooperation, and even affect employees’ health (Abidin et al., 2020; Falk et al., 2018).

Compensation models that govern pay, amongst other things, vary in what they reward: the demands of the job, the dynamics of the labour market, or the qualities or output of the individual employee. These models reflect different understandings of fairness and apply different criteria for determining pay. Table 1 presents an overview of three commonly used compensation models: job-based, market-based, and person-based models.

Table 1.

Job-Based, Market-Based, and Person-Based Compensation Models

Model	Job-based	Market-based	Person-based
Criteria for compensation	Requirements and characteristics of the job	External market value of the job	Individual competencies, skills, productivity, or effort
Primary instruments	(Formal) job evaluation methods	Market pricing, salary benchmarking	Performance appraisals, skill assessments, individual benchmarking
Common contexts	Large or complex organizations with structured work	Competitive or rapidly changing sectors with unstructured work	Innovation-driven or research-oriented environments
Strengths	Promotes equity and transparency; supports equal pay compliance	Competitive pay; enables rapid market responsiveness	Tailored compensation; recognizes individual development and value
Limitations	May be costly, rigid, or bureaucratic	Risk of wage inflation; potential erosion of fairness principles	High implementation cost; vulnerable to bias; consistency challenges

Inspired by Armstrong (2018) and Milkovich et al. (2023).

Each model prioritizes different criteria. What counts as a fair salary in one model may not be considered fair in another. For example, a job that requires few specialized skills but is in high demand due to labour shortages would be valued differently in the different models. A job-based model would consider the limited skill requirement and assign a relatively modest compensation. A market-based model, in contrast, would emphasize scarcity and demand, likely resulting in a higher pay level to remain competitive. Among the different approaches to compensation, job-based models remain widely used in organizations across sectors (Van de Glind et al., 2025). The rise of job-based models dates back to the end of World War II and is linked to the emergence of collective labour agreements (Oettinger, 1964). These systems are designed to foster internal equity by systematically evaluating jobs, and to help control costs. Although job evaluation systems remain widely used to support equitable pay, we know relatively little about whether the criteria they rely on are (still) perceived as fair by society.

Justice Theories and the Perception of Fair Pay

At a more conceptual level, several theoretical traditions offer insights into what fair compensation is and how it can be achieved. These frameworks are widely used in fields such as human resource management, labour economics, industrial relations, and organizational psychology. Here, we briefly address three perspectives: distributive justice (including equity theory), procedural justice, and interactional justice.

First, distributive justice concerns the perceived fairness of outcomes, such as salaries, bonuses, and other benefits. According to equity theory of motivation (Adams, 1963), individuals judge fairness by comparing the ratio of their own inputs (e.g. effort, experience, skills) to outputs (e.g. pay, recognition) with that of others. What counts as fair depends on implicit distributive norms within a group or society, such as equality (equal outcomes), equity (proportional inputs and outputs), need (allocations based on individual need), power (allocations based on influence), or responsibility (greater rewards for heavier burdens) (Le Garrec, 2018).

Second, procedural justice focuses on the fairness of the decision-making process, including transparency, consistency, and the opportunity to have a voice in compensation decisions (Scott, 2018).

Third, interactional justice emphasizes how these processes are experienced by individuals. It covers both interpersonal justice, that is, the degree of dignity and respect shown, and informational justice, the adequacy and honesty of explanations provided about pay decisions (Scheller & Harrison, 2018). Together, these three perspectives capture compensation fairness as: (a) fairness of outcomes, (b) fairness of procedures, and (c) fairness in communication and interpersonal treatment.

The Fairness of the Criteria Used in Job Evaluation Methods

Given their reputation and widespread adoption, this article focuses on the criteria used in job evaluation. Job evaluation methods, whether point-factor systems, in which jobs receive a relative value based on compensable factors, classification schemes, or other formal techniques, are widely seen as evidence-based tools to support transparent and equitable job-based pay structures. They are recognized as instruments that help ensure equal pay and procedural fairness (Armstrong, 2018; Marcinkowski et al., 2024; Oelz et al., 2013). They are used towards achieving fairness of outcomes, that is, distributive justice. Existing systems are developed within specific historical, organizational, and economic contexts. Prior research has shown that job evaluation practices are influenced by societal changes and therefore should be periodically reassessed to remain aligned with current norms and values (Armstrong, 2018; Figart, 2000). This is particularly relevant in light of the recent Directive (EU) on pay transparency which obliges employers to disclose pay structures, justify pay differentials, and use objective and gender-neutral criteria in pay setting. Formal job evaluation is recommended as a reliable tool to achieve these goals. Therefore, it is especially important to ensure that systems produce fair outcomes. However, recent research shows that scholarly attention to job evaluation has significantly declined since 2010, with few recent empirical studies available (Van de Glind et al., 2025). This decline has left a gap in understanding whether the criteria used in job evaluation are still relevant, legitimate, and sufficient in the eyes of the broader public. This study aims to that gap, by exploring answers to the questions: what criteria are used to determine the relative value of jobs within these systems; and are these criteria relevant and sufficient in light of current societal norms?

Methods

We applied a method known as Participatory Value Evaluation (PVE) to explore public preferences concerning compensation fairness. The essence of the PVE method is that participants are presented with several policy options and a budget constraint. Next, they are asked which options should be chosen subject to this budget constraint (Mouter et al., 2022). PVE has previously been applied to investigate public opinion in sectors such as healthcare, energy, and transport (de Ruiter et al., 2025; Mouter et al., 2022; Mulderij et al., 2021), but never in the domain of job compensation.

In the PVE task of this study, participants were asked to distribute an additional €1000 in salary across five different jobs within the same income range, and subsequently explain their decisions for each job. Each participant completed two sets of five jobs resulting in 10 choices and supporting arguments per participant. The PVE was embedded in an online survey consisting of three parts: (1) sociodemographic questions, (2) an introduction to the PVE choice task, and (3) the task itself. Each job was displayed by title, and participants could access a plain-language description by clicking on an information icon. Information about the salary paid in the different jobs was not provided, but participants were explicitly informed that the jobs received a similar salary. Participants submitted their justifications in free-text boxes, one per job, with the open-ended question about the choices they made in their distribution of extra salary.

The comparative nature of the task was expected to elicit richer arguments than standard rating or ranking exercises. A pilot test with 149 participants showed minor comprehension issues in the task instructions, which were subsequently clarified in the final version of the survey. No changes were made to the job sets or survey structure as a result of the pilot.

Selection of Jobs

We applied the following criteria for jobs to be included in the survey. To stimulate a diverse range of arguments, we included jobs that varied along three aspects: income level (Dutch monthly salary levels for full time work, for example, low: < €2500, middle: €2500–4000, high: > €4000), sector, and known job attributes such as responsibility, educational requirements, physical and emotional demands, risk, and irregular working hours. Candidate jobs were excluded if the job (1) was considered not commonly recognizable, (2) is often target of ridicule or controversy (e.g. hospital clowns), (3) is typically performed by self-employed individuals rather than employees. We oversampled healthcare jobs (21 out of 45) because we were particularly interested in arguments used to give those jobs more or less additional salary.

In total, nine sets of five jobs were included in the survey. We designed the subsets so that only salary and sector changed between them, while the other known job attributes varied within each subset. This way, we could create nine sets of jobs in which the variation between jobs was primarily due to the previously mentioned known job attributes. Also, in order to reduce the risk of a bias towards healthcare jobs, we created sets with only healthcare jobs, with only non-healthcare jobs and with both healthcare and non-healthcare jobs combined. In the final PVE, each respondent was asked to complete 2 sets of five jobs.

Sample and Data Collection

Data were collected between September 26 and October 11, 2022. In total, 3,734 individuals completed the survey (Table 2). The survey platform recorded each respondent’s salary distribution decisions and their written justifications for each choice that they made. The open-text explanations provided the main source of qualitative data. For the aim of this study, all relevant groups were represented substantially. Participants were recruited via a commercial panel agency and received a standard incentive. Participation was voluntary and anonymous. Ethical approval for this study was not required under Dutch law.

Data Analysis

Demographic data were analysed descriptively. Open-text responses were analysed using the qualitative content analysis approach described by Hsieh and Shannon (2005), facilitated by NVivo version 12 (2018). This method suited the exploratory nature of the study and allowed the coding process to remain closely tied to participants’ own words. Monetary data were analyzed descriptively.

Initial familiarization with the open-text data began during the pilot phase. Each of the eight coders individually read a sample of responses and generated preliminary codes. The team then met to compare interpretations and establish a shared understanding. This collaborative process led to the development of an initial coding framework. The remaining data were divided among team members, who continued coding iteratively while remaining alert to new themes.

Once the coding framework had stabilized, seven trained research assistants were added to the team. They were briefed to follow the codebook but also encouraged to flag new or unclear cases, which were then discussed in the group until consensus was reached. Coding continued until thematic saturation was reached. In total, 27,075 out of 41,075 responses (67.4%) were coded.

Table 2.

Demographic Characteristics of the Sample

Characteristic	% (n)	Dutch citizens aged >15 years (Centraal Bureau voor de Statistiek, 2021)
Total	100% (3734)	100%
Gender
Male	43.2% (1610)	49.5%
Female	56.7% (2114)	50.5%
Other/Prefer not to say	<0.1% (7)	Unknown
Age
<25 years	12.2% (456)	14.8%
25–44 years	45.5% (1696)	29.4%
45–64 years	28.7% (1072)	32.7%
≥65 years	13.6% (506)	23.1%
Education
Primary and lower secondary education	16.4% (611)	29.0%
Upper secondary and intermediate vocational education	35.3% (1318)	36.6%
Higher education	48.3% (1801)	34.4%

Development and Structure of the Codebook

The development of the codebook involved a series of interpretive decisions aimed at ensuring consistency and analytical depth. One challenge was the brevity of many responses. While some participants provided detailed justifications, others used short phrases such as ‘tough job’, which could refer to different aspects of the work – for example, the physical or emotional demands. To address this ambiguity, we formulated coding guidelines.

If the job in question was clearly physically strenuous (e.g. a mover), the response was coded as ‘physically demanding’. When the intended meaning was unclear, multiple codes were assigned (in this case: both ‘physically demanding’ and ‘emotional labor’) to avoid overinterpreting limited data.

To capture the direction of each argument, all codes were marked with a ‘(+)’ or ‘(−)’, depending on whether the argument was used to justify allocating more or less additional salary. For instance, a response such as ‘you don’t need a lot of education for this’, in combination with a lower allocation, was coded as ‘education necessary (−)’, whereas a comment like ‘needs a lot of knowledge’, combined with a higher allocation, was coded as ‘education necessary (+)’.

If responses provided more detailed or nuanced reasoning, an additional layer of subcodes was introduced. For example, a participant who wrote, ‘tough job emotionally, because you hear many personal stories’, offered more than a generic claim. In this case, the response was coded not only as ‘emotional labor’ but more specifically as ‘emotionally demanding’. This subcode was developed to distinguish exposure to suffering, trauma, or moral complexity from the relational and reflective elements of emotional labor, such as managing emotions, practicing empathy, and engaging in self-reflection. The latter category was coded as ‘emotional skills’. Not all parent codes required subcodes, but they were added when greater specificity was needed to be able to capture the participants’ argumentation. Figure 1 shows a section of the codebook and how subcodes could be added if the responses were nuanced enough. The full codebook is available upon request.

Figure 1.

A section of the codebook showing parent- and sub-codes with specifications

Some responses combined societal and market-based reasoning, such as highlighting both a profession’s essential role as well as current shortages, we prioritized the code that most closely aligned with the explicit justification. For example, if a participant emphasized the societal importance of teaching alongside a comment on labor shortages, and the stated reason focused on ‘importance for the future of children’, the response was coded under ‘societal importance’ rather than ‘labor market’.

Finally, we considered the responses of participants who divided the €1000 equally across all five jobs. These were only interpreted as reflecting egalitarian values if supported by explicit reasoning, such as ‘everyone deserves the same’. This helped to ensure that our coding remained grounded in participants’ stated justifications rather than our assumptions about their intentions.

Final Codebook Structure

The final coding structure grouped arguments into two main categories: job-related arguments, and market-based arguments. Table 3 provides an overview of the two categories, their subcategories, example arguments, and assigned codes.

Table 3.

Final Categories, Subcategories, Example Arguments, and Assigned Codes

Category	Subcategory	Example argument	Code assigned
Job-related arguments	a. Requirements of the job	‘Had to study a lot’	Education needed (+)
	b. Work environment	‘It’s nice to be outside all day’	Weather (−)
		‘Here you run the risk of being attacked by an addict’	Risks (+), subcode: Risk of personal injury (+)
	c. Consequences of the job	‘Lots of things can go wrong during surgery’	Consequences of mistakes (+), subcode: Risk of human harm (+)
	d. Social contribution	‘Children are the future, so this work is very important’	Societal importance (+)
Market arguments	a. Labor market position	‘Hard to find personnel in this sector’	Labor market (+)
Market arguments	b. Job image	‘It’s quite hard work and very important, but not very appreciated’	Job is underrated (+)

(Full Codebook available in Appendix A.)

The final categories were compared to the criteria used in literature on designing job evaluation systems and to the criteria used in well-known and openly accessible job evaluation systems. This is discussed in the Results section.

Results

Monetary Data

In general, participants were more likely to allocate more additional salary to public sector jobs than to private sector jobs, but the difference was small, EUR 24,-. Appendix B provides an overview of how the additional salary was distributed in each individual set of jobs. Because participants were randomly assigned two of the nine sets of jobs, all with a different composition in terms of salary level and healthcare/non healthcare jobs, it is not possible to generalize their choices and provide a total rank order of the jobs.

Job-Related Arguments

Within the two categories, we identified four subcategories. The four subcategories can be understood as follows. Arguments about the job’s requirements mention the educational requirements, necessary skills, and competencies that are needed for the job. Some of these requirements may be mandated by law, such as certain diplomas or training. They are legally necessary to perform the job. We included arguments like ‘had to study a lot’ (coded as ‘education needed’) in this category. Other requirements are not mandatory in a legal sense, but are judged by the respondent as needed for the job (to a greater or lesser extent). For example, one respondent wrote, ‘only fit for a certain kind of people, the righteous kind’, as an argument to give a judge extra salary, alluding to the need to possess integrity for this position (coded as ‘integrity and trustworthiness’).

The subcategory of arguments about the work environment contains all arguments concerning the circumstances in which the job is performed. Here, we included arguments about the weather conditions and exposure to hazardous materials, dangerous situations, and dirt, foul smells etcetera. People would give arguments that emphasized how much harder such work environments made the job. Interestingly, there were arguments about work environments that were used both for giving more, as well as less, extra salary. Take the postal worker, for instance. Some participants allocated extra money because they observe that postal workers have to work in all weather conditions, or less extra money, because ‘it’s nice to be outside all day’. The subcategory of the work environment also includes arguments that mention the workload and the hustle and bustle associated with the job.

The third subcategory comprises arguments about the job consequences. Here, we mean consequences in a broad sense. For example, there are arguments about the consequences of a job for the worker themselves: the degree of physical or mental burden of a job, or the extent to which the worker’s private life is affected, for instance, by irregular working hours or nightshifts. But we also included arguments about the potential consequences of errors, mistakes, and faults in this category. An organization could, for instance, be harmed or damaged as a result of a worker’s actions (because of reputational or material damage). Or, it is the life of other people, or the quality of it, that may be jeopardized by a worker’s actions, such as mistakes made by a doctor or nurse.

We clustered all arguments about the job’s contribution to society in the subcategory social contribution of the job. Oftentimes, participants would either implicitly or explicitly compare the job’s contribution to the contribution of the other jobs listed in the task. Participants more often attributed additional salary to jobs that have, in their view, a greater societal value, in contrast to jobs with a smaller or even absent added social value. Some participants linked their argumentation to current societal developments. For instance, participants who granted extra salary to the agricultural scientist argued that ‘given the current nitrogen crisis [in The Netherlands], we need scientists to help find solutions’. Another example is the argument of ‘our society is aging, so this group [of workers] will become more important’.

Market Arguments

The arguments in the category market arguments represent a job’s value based on external factors. In this category, we distinguished two subcategories: shortage of workers (or labor) and image. This latter subcategory holds the arguments in which participants expressed how, in their view, the job in question was currently either over- or under-rated.

In the subcategory shortage of workers (or labor), participants distributed extra salary to a job because there is a shortage of workers in that job (e.g. primary school teachers). The nature of the scarcity or need for jobs (or labor) often had to do with topics that also played a role in the subcategory social contribution. For instance, if additional salary was used as a means to incentivize people to choose or stay in such jobs: ‘they are needed, so make it more attractive’ as a participant wrote about the job of care aid.

The subcategory image includes arguments in which participants expressed that the job in question was either over- or under-rated. These arguments could also refer to aspects related to the job. This is, for instance, the case in ‘They already make plenty, [while] it’s just about putting numbers in a computer program’, about the job of mortgage advisor. Here, the respondent indicates that you do not need a lot of skill to be able to do this job (a job requirement). But because they also refer to the relative worth of the job, that is, they make plenty for what they do in comparison with other workers, we have put this type of response in the subcategory image. It is not only that the worker does not need to be highly trained that they do not receive extra salary from the respondent – it is because the respondent thinks mortgage advisors already make more than they actually deserve. The participant would rather see the other jobs in the set get more additional salary.

Comparison with Criteria Currently Used in Job Evaluation Systems and Guidelines

Table 4 shows a comparison between the themes from our qualitative analysis with criteria from well-known job evaluation systems and guidelines: the Hay Chart-Profile Method of job evaluation (developed by the Hay group, now part of Korn Ferry), NHS job evaluation scheme or guidelines for such systems (International Labour Organisation, Armstrong). This comparison illustrates where public reasoning overlaps with, extends beyond, or contrasts with the contents of existing formal job evaluation systems. It shows that the arguments made by participants in this study often, but not always, align with the criteria used in formal systems, particularly regarding required qualifications, responsibilities, and job strain.

Table 4.

Comparison of Arguments of the Public With Job Evaluation Factors/Criteria

Reasoning in this study	Example arguments	ILO^a	Hay evaluation method^b	Armstrong^c	NHS^d
Job-related arguments
Job requirements: Educational	‘Had to study a lot’	Qualifications and skills (including academic, vocational, informal)	Know-how	Skills, knowledge, expertise	Knowledge, training and experience
	‘Technical skills and education needed’				Communication and relationship skills
	‘Compared to the rest, least education needed’				Physical skills
	‘No extra education needed’
Job requirements: Complexity and judgement	‘This requires a lot of judgement’	—	Problem solving	Judgement, analysis, decision-making	Freedom to act
	‘A lot of IT knowledge required’				Analytical and judgemental skills
	‘You need specialised knowledge’
	‘Not too complex, students can also do it’
Potential consequences of the job: Emotional and physical demands	‘Tough job emotionally’	Effort (emotional, mental, physical)	—	Emotional/environmental demands	Physical and mental effort
	‘You have to be able to stay true to yourself and let go of all these stories’
	‘Jobs that are hard on the body should be compensated better so they can retire sooner’
	‘My son works in a clothing store. He walks 15 km a day. It’s really tough’
The job’s social contribution	‘Children are the future’	—	—	—	—
	‘This is not essential’
	‘A soldier stands up for their people and should therefore get more pay’
	‘They take care of our environment’
Work environment	‘You could get attacked by an addict’	Working conditions (physical, psychological)	—	Environmental demands	Working conditions
	‘Risk-free job’
	‘Jobs nobody wants because it’s so dirty’
	‘High workload in healthcare’
Market-related arguments
Labor market shortages	‘Hard to find personnel in this sector’	—	—	—	—
Labor market shortages	‘If you’re really good at something and there are not many other people who can do it, you can ask for higher pay’
Image or perceived prestige	‘They already make plenty for what they do’	—	—	—	—
	‘Could be done by volunteers’
	‘If you choose this job it is likely a calling’

^a(Chica, 2008).

^b(University of Waterloo, n.d).

^c(Armstrong, 2018).

^dNHS Employers (2024).

Discussion

Summary of Findings

Job evaluation is a widely used tool in organisations and sectors to support fair and consistent pay. To ensure it continues to produce fair outcomes, its underlying criteria must be examined periodically. This study examined public perceptions of compensation fairness and compared them with criteria commonly used in job evaluation systems. More than 3,700 Dutch citizens participated in an innovative online experiment, in which they allocated additional salary to various jobs and explained their reasoning.

Our analysis showed that the arguments people used were largely aligned with criteria embedded in job evaluation systems. Specifically, we identified four categories of job-related arguments: (a) the job requirements, (b) the work environment, (c) the potential job consequences, and (d) the job’s social contribution. The first three categories are largely covered by existing criteria. However, the social contribution of a job, its perceived value to society, was frequently cited by participants as a justification for higher pay but is not currently included in most job evaluation frameworks. We also identified two categories of market-related arguments: (a) perceived labor shortages, and (b) the occupational image or prestige of a job. Participants considered both to be valid grounds for increased compensation, although these arguments are typically intentionally excluded from job evaluation systems, as they do not pertain to the characteristics or demands of the work itself.

Interpreting the Results

To our knowledge, this is the first study to empirically examine how well the foundational principles of job evaluation systems align with public perceptions about fair compensation. That the majority of public arguments align with current criteria suggests that these systems remain largely robust and socially grounded in The Netherlands. This is important, as job evaluation frameworks play a stabilizing role in salary structures in organizations or across entire sectors. Adapting or changing these frameworks has potentially wide-reaching effects for employers, employees, and labor relations. From this perspective, our findings support the idea that job evaluation continues to be a valuable tool.

At the same time, our results suggest that important societal concerns may not be fully captured by current systems. Social value and labor market dynamics, particularly shortages and perceptions of occupational worth, were recurring themes in participants’ reasoning. The inclusion of such arguments in people’s evaluations suggests that notions of fairness are not limited to what happens within the job itself but also include external and societal considerations. Our findings suggest that greater reflection is needed on how job evaluation systems account for aspects of work that are increasingly viewed as valuable, but that are currently undercompensated or neglected. At the same time, it is important to recognise that public perceptions of fairness primarily constitute a normative or theoretical point of reference, rather than a direct input, in the design and evaluation of job evaluation systems. Such systems operate within organisational and institutional contexts shaped by internal equity principles, collective agreements, budgetary constraints, and strategic objectives. As a result, considerations that resonate strongly in public reasoning, such as labour market scarcity or occupational image, may not be compatible with the underlying logic or objectives of job-based pay systems, even when they are perceived as relevant to fair compensation.

Importantly, job evaluation typically constitutes only one component of broader compensation systems, within which other instruments may be used to respond to labour market pressures without altering the evaluation of the job itself.

The findings should be interpreted in light of the Dutch context. The Netherlands is known for its egalitarian values and relatively small income differences, supported by collective bargaining traditions and a cultural preference for equality and moderation (Hofstede Insights, n.d). These values may help explain why participants focused on fairness, contribution, and proportionality rather than competition or exceptional individual performance. At the same time, Dutch citizens often combine egalitarian norms with meritocratic beliefs that reward education and effort (Batruch et al., 2022; Van Oorschot, 2000). This mix of solidarity and merit helps to understand the balance participants struck between valuing equality and recognizing differences in responsibility or skill. This specific institutional and cultural configuration should be taken into account when considering the applicability of these findings beyond the Dutch case. Hence, rather than assuming direct generalizability of the substantive findings, the question of how public reasoning about fair pay varies across countries may be viewed as an empirical one. In contexts with broadly similar institutional arrangements and value orientations, such as countries with strong collective bargaining traditions and relatively compressed wage structures, similar patterns of reasoning may be expected. In contrast, in settings where wage-setting is more individualized or where meritocratic or hierarchical principles play a more dominant role, public arguments about fair compensation may differ substantially.

The participatory value evaluation approach applied in this study offers a way to systematically examine such differences. Replicating this study in other institutional contexts would allow for comparative insight into how societal norms, labour market institutions, and compensation systems interact in shaping perceptions of fair pay.

Our study is particularly relevant in light of the recent Directive (EU) on pay transparency which obliges employers to disclose pay structures, justify pay differentials, and use objective and gender-neutral criteria in pay setting. The directive explicitly identifies analytical job evaluation methods as a means to demonstrate compliance with the principle of equal pay for work of equal value. Our study contributes to this policy context by providing an up-to-date and empirically grounded perspective on what people believe should count when it comes to determining the value of a job. The strong alignment between existing job evaluation systems and public reasoning offers reassurance that these systems are still perceived as legitimate. However, the minimal attention to the criterion of social contribution in job evaluation frameworks raises the question regarding the extent to which such systems are sufficiently equipped to reflect what people currently consider fair, particularly in sectors where gender pay gaps are persistent. This may be especially relevant in public sectors such as healthcare, education, and social work, where pay gaps are often connected to the undervaluation of work traditionally performed by women (Grimshaw & Rubery, 2007).

Our findings also resonate with research on occupational prestige and value. While occupational rankings are typically stable across time and cultures, with surgeons, judges, and airline pilots consistently at the top, there is increasing recognition that social value and occupational prestige are distinct concepts. Newlands (2024) found that healthcare occupations dominated rankings of social value, whereas prestige-based rankings often favoured other professions. In our study, participants consistently valued jobs in healthcare, education, police, defence, and public infrastructure for its societal importance. As such, this study adds empirical support to emerging calls for a review of existing job evaluation systems and their criteria, especially where public sector and essential services are concerned.

At present, job evaluation systems are typically grounded in meritocratic principles that emphasise knowledge, skills, and responsibilities. This study’s findings reveal a plurality of distributive justice perspectives. This diversity indicates that people draw on multiple, and sometimes conflicting, conceptions of fairness. For job evaluation systems, this underscores the growing importance of transparency about which justice principles are applied and why.

As with any study on public opinion, our findings represent just one input into evaluating job evaluation systems. Other considerations must include employer needs, feasibility, strategic priorities, and ongoing changes in the nature of work. For instance, while responsibility remains a key criterion, how it is exercised may evolve due to digitalisation or organisational shifts and changed working practice. This underscores the need to not only examine which aspects are used in job evaluation systems but also how they are defined, operationalised, and weighted.

We recommend future research to explore: how criteria in job evaluation systems are defined and operationalised; whether the relative weight of the criteria is still justified; and how sectoral debates about fairness may influence compensation reforms. We also encourage a broader debate about the justice principles that guide the design of job evaluation systems, and whether these are still in step with current or foreseen circumstances. Lastly, it would be interesting to explore public preferences about this topic in other societies to be able to compare and contrast values.

Contributions of This Study

This study offers three key contributions. First, it provides a current empirical inventory of public reasoning about job compensation across a wide spectrum of occupations. This dataset can serve as a reference point for assessing the continued relevance of criteria of job evaluation systems in all kinds organisations and sectors. Second, it invites a broader debate about whether and how to incorporate social value and market-based reasoning into job evaluation. This may help organizations align their pay structures with both regulatory requirements and evolving societal norms. Understanding how people think about the value of jobs, particularly in terms of its societal relevance and labor market context, can also inform future adjustments to job evaluation systems. This is especially relevant in the public sector, where shortages are pressing and legitimacy concerns are high (Michaeli et al., 2024; Symeonidis et al., 2025; Wilson & MIles-Johnson, 2024). This creates opportunities for organisations that use a job evaluation system as the foundation of their compensation model, and have trouble attracting or sustaining personnel. The findings indicate that people may understand and agree that for some jobs, ones that are in high demand or that may have a significant social contribution, it is fair to pay more. This knowledge creates an opportunity for custom work in compensation: for instance to install a temporary increase in pay for some jobs. Third, this study introduced a novel qualitative experiment to the field of compensation research. We hope it inspires further methodological innovation and comparative studies on job evaluation.

Conclusion

This study examined the arguments that Dutch citizens use when asked to allocate additional salary across a range of jobs, and compared their arguments with the criteria currently used in job evaluation systems. It shows that existing job evaluation frameworks remain largely aligned with public reasoning about the requirements, demands, and responsibilities of work. However, the participants also mentioned social contribution and the labour market context as potential grounds for fair pay. As labour shortages persist and debates about essential work intensify, these results invite reflection on how compensation systems can remain both just and credible. We further recommend revisiting the operationalisation and weighting of criteria in job evaluation and initiating broader public and institutional discussions on the evolving values that underpin fair compensation.

Future research could explore how criteria in job evaluation systems are defined and operationalised; whether the relative weight of the criteria is still justified; and how sectoral debates about fairness may influence compensation reforms. We also encourage a broader debate about the justice principles that guide the design of job evaluation systems, and whether these are still in step with current or foreseen circumstances. Lastly, it would be interesting to explore public preferences about this topic in other societies to be able to compare and contrast values.

Supplemental Material

Supplemental Material - What Counts when it Comes to Salary? A Qualitative Analysis of the Arguments Dutch Citizens Use when Distributing Salary

Supplemental Material for What Counts when it Comes to Salary? A Qualitative Analysis of the Arguments Dutch Citizens Use when Distributing Salary by Jael Bootsma, Irene van de Glind, Roos Mulder, Agnes Akkerman, Evelyn Finnema, Lex Heerma van Voss, Tom Geijsen, Niek Mouter, Geertje van de Ven in Compensation & Benefits Review

Supplemental Material

Supplemental Material - What Counts when it Comes to Salary? A Qualitative Analysis of the Arguments Dutch Citizens Use when Distributing Salary

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Jael Bootsma

Irene van de Glind

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Jael Bootsma, PhD is senior researcher in the Department of Research and Development of FWG Foundation for job evaluation in healthcare.

Irene van de Glind, PhD works as a senior researcher at the Department of Research and Development of FWG Foundation for job evaluation in healthcare.

Roos Mulder, MSc is a PhD Candidate involved in the Strategic Human Resource Management study program at Utrecht University.

Agnes Akkerman works at the Institute for Advanced Labour Studies / Hugo Sinzheimer Institute (AIAS-HSI) at the University of Amsterdam, as professor of Regulation of Labour and head of the department of Labour Law.

Evelyn Finnema is professor in Nursing Science and Education at the University of Groningen and member of the Executive Board NHL Stenden University of Applied Sciences.

Lex Heerma van Voss, PhD is a research Fellow at the International Institute of Social History in Amsterdam.

Tom Geijsen, MSc is a senior researcher at Populytics, focusing on the evaluation of government projects.

Niek Mouter, PhD is associate professor at Delft University of Technology, focusing on the evaluation of government projects.

Geertje van de Ven, PhD is senior researcher and head of the Department of Research and Development of FWG Foundation for job evaluation in healthcare.

References

Abidin

N. Z.

Pangil

Othman

S. Z.

(2020). Pay inequity and employee outcomes: The mediating role of perceived organizational justice. International Journal of Academic Research in Business and Social Sciences, 10(6), 1054–1064. https://doi.org/10.6007/IJARBSS/v10-i6/7353

Adams

J. S.

(1963). Toward an understanding of inequity. The Journal of Abnormal and Social Psychology, 67(5), 422–436. https://doi.org/10.1037/h0040968

Armstrong

(2018). Armstrong’s job evaluation handbook: A guide to achieving pay fairness and transperancy in pay and reward. Kogan Page Publishers.

Batruch

Jetten

Van de Werfhorst

H. G.

Darnon

Butera

(2022). Belief in school meritocracy and the legitimation of social and income inequality. Social Psychological and Personality Science, 14(5), 632–645. https://doi.org/10.1177/19485506221111017

Belcher

D. W.

(1979). Compensation administration (3rd ed.). Prentice-Hall: Englewood Cliffs, NJ.

Centraal Bureau voor de Statistiek . (2021). Bevolking; kerncijfers. https://opendata.cbs.nl/statline/portal.html?_catalog=CBS&_la=nl&tableId=37296ned

Chica

M. T.

(2008). Promoting equity: Gender neutral job evaluation for equal pay: A step-by-step guide. International Labour Office (ILO).

de Ruiter

Tuit

Mouter

(2025). Beyond face validity: Assessment of a participatory value evaluation for an institutionalized transport conflict. Transport Policy, 171, 114–127. https://doi.org/10.1016/j.tranpol.2025.05.024

Falk

Kosse

Pinger

(2018). Mentoring and school performance: Evidence from a randomized field experiment. The Review of Economic Studies, 85(3), 1800–1832. https://doi.org/10.1093/restud/rdx046

10.

Figart

D. M.

(2000). Equal pay for equal work: The rol of job evaluation in an evolving social norm. Journal of Economic Issues, 34(1), 1–19. https://doi.org/10.1080/00213624.2000.11506241

11.

Grimshaw

Rubery

(2007). Undervaluing women’s work. Manchester: University of Manchester.

12.

Hofstede Insights . (n.d.). Country comparison: Netherlands. https://www.hofstede-insights.com/country-comparison/the-netherlands/

13.

Hsieh

Shannon

S. E.

(2005). Three approaches to qualitative content analysis. Qualitative Health Research, 15(9), 1277–1288. https://doi.org/10.1177/1049732305276687

14.

Le Garrec

(2018). Fairness, social norms and the cultural demand for redistribution. Social Choice and Welfare, 50(2), 191–212. https://doi.org/10.1007/s00355-017-1080-6

15.

Marcinkowski

Butnaru

Rabrenovic

(2024). Salary systems in public administration and their reforms. SIGMA Papers, (No. 63). https://doi.org/10.1787/8f08a005-en

16.

Michaeli

D. T.

Michaeli

J. C.

Albers

Michaeli

(2024). The healthcare workforce shortage of nurses and physicians: Practice, theory, evidence and ways forward. Policy, Politics, & Nursing Practice, 25(4), 216–227. https://doi.org/10.1177/15271544241286083

17.

Milkovich

G.T.

Newman

J.M.

Gerhart

(2023). Compansation (14th ed.). New York, NY: McGraw-Hill Education.

18.

Mouter

Jara

K. T.

Hernández

J. I.

Kroesen

de Vries

Geijsen

Kroese

Uiters

de Bruin

(2022). Stepping into the shoes of the policy maker: Results of a participatory value evaluation for the Dutch long term COVID-19 strategy. Social Science & Medicine, 314(115430), 115430. https://doi.org/10.1016/j.socscimed.2022.115430

19.

Mulderij

L. S.

Hernández

J. I.

Mouter

Verkooijen

K. T.

Wagemakers

(2021). Citizen preferences regarding the public funding of projects promoting healthy body weight among people with a low income. Social Science & Medicine, 280, 114015. https://doi.org/10.1016/j.socscimed.2021.114015

20.

Newlands

G. L. C.

Lutz

(2024). Mapping the prestige and social value of occupations in the digital economy. Journal of Business Research, 180, 114716. https://doi.org/10.1016/j.busres.2024.114716

21.

NHS Employers . (2024). NHS job evaluation handbook. https://www.nhsemployers.org/publications/nhs-job-evaluation-handbook

22.

Oelz

Olney

Tomei

(2013). Equal pay: An introductory guide. International Labour Office (ILO).

23.

Oettinger

M. P.

(1964). Nation-wide job evaluation in the Netherlands. Industrial Relations: A Journal of Economy and Society, 4(1), 45–59. https://doi.org/10.1111/j.1468-232X.1964.tb00911.x

24.

Scheller

E. M.

Harrison

(2018). Ignorance is bliss, or is it? The effects of pay transparancy, informational justice and distributive justice on pay satisfaction and affective commitment. Compensation & Benefits Review, 50(2), 65–81. https://doi.org/10.1177/0886368719833215

25.

Scott

(2018). Pay communications and fairness: An employee perspective. Compensation and Benefits Review, 50(1), 5–20. https://doi.org/10.1177/0886368718809774

26.

Symeonidis

Guberman

Cooper

(2025). Addressing teacher shortages in an international context: Implications for the quality and status of teacher education. European Journal of Teacher Education, 48(1), 1–8. https://doi.org/10.1080/02619768.2025.2447968

27.

University of Waterloo

H. R.

(n.d.). Hay evaluation method. https://uwaterloo.ca/human-resources/support-managers/compensation-information-managers/hay-evaluation-method

28.

Van de Glind

Mulder

Akkerman

Van der Biezen

Bootsma

Finnema

Van de Ven

Mouter

van Rooij

(2025). Mapping the literature on job evaluation: A scoping review. Compensation and Benefits Review, 57(1), 24–46. https://doi.org/10.1177/08863687241279592

29.

Van Oorschot

(2000). Who should get what, and why? On deservingness criteria and the conditionality of solidarity among the public. Policy & Politics, 28(1), 33–48. https://doi.org/10.1332/0305573002500811

30.

Wilson

J. M.

MIles-Johnson

(2024). The police staffing crisis: Evidence-based approaches for building, balancing, and optimizing effective workforces. Policing: A Journal of Policy and Practice, 18(paae044), paae044. https://doi.org/10.1093/police/paae044

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.53 MB

0.31 MB

0.00 MB