The Influence of Total Compensation on Job Satisfaction*

This study aims to examine the effects of several total compensation factors on job satisfaction. This cross-sectional and confirmatory research has collected data from a sample of 246 employees, including lower-level workers, technicians, and middle managers in manufacturing and service industries in Spain, mainly medium- and large-sized companies in the private sector. Data analysis includes reliability and validity tests of psychometric scales and hypotheses testing using multi-group Structural Equations Modelling (SEM). The study results show that supervisor support, career development, and work-life balance are strongly and positively related to job satisfaction. However, pay satisfaction is positively associated with job satisfaction only for the group of employees who are 40 years of age or younger. Managers must conceptualize these factors as part of total compensation and promote work flexibility, career development, and supervisor support policies. The results add to the small literature investigating the growing field of total compensation and its outcomes for job satisfaction and employee performance.

Traditional compensation systems, based on fixed and variable rewards, were common during the second half of the 20 th century and gradually incorporated other benefits such as medical insurance, restaurant tickets, or transport rewards, among others. However, these compensation systems represent an old paradigm in most evolved countries, as these are no longer able to attract and retain new employees, more eager for positive feedback, teamwork climate, and good communication, as well as to fully develop themselves in their workplace (Cahill & Sedrak, 2012). Employees in the 21 st century tend to aspire to other intangible benefits that are gaining relevance in the valuation of the workplace and that have begun to be termed "Total Compensation" (Hoole & Hotz, 2016).
Total Compensation (TC) includes intangible factors such as the quality of management (Griffin et al., 2001), the company values (Bustamante et al., 2021), communication transparency (Nikolić et al., 2013), the organization itself (Azanza et al., 2013) or an explicit HR strategy setting clear rules for the relationships between the company and its employees (Scott & McMullen, 2010). Some authors (e.g., Ulrich & Smallwood, 2003) have studied compensation elements termed "non-financial rewards" such as customized deals tied to competence and performance, high levels of autonomy, effective feedback, or the opportunity to complete an entire task (end to end process). Other authors begin to address this paradigm shift (Laloux, 2016), highlighting the need to transform companies into more open organizations (Carney & Getz, 2018), where professional and personal development are compatible and where the workspaces are designed to increase productivity as well as to develop people's abilities fully.
Some of the most increasingly relevant factors in today's workplace are the ability to enjoy a healthy work-life balance while maintaining professional prospects, receiving support from one's supervisor, and earning a decent salary (González et al., 2017). This is especially important for employees and middle managers, who represent the largest group in companies and organizations. Despite the importance of these factors, there is a lack of sufficient research on them and, more specifically, their influence on job satisfaction (Dinkin, 2015).
Knowing the influence of these factors on job satisfaction is critical for companies and organizations since they are intangible -yet, highly valued-compensation in scenarios of stagnating wages due to global competition (Recio, 2001). In many cases, wages have even been reduced due to the fairly recent financial crisis, which created a growing wage gap due to the emergence of low-paid jobs in Spain (Fernández Romero et al., 2013).
The most recent studies published on job satisfaction in Spain were conducted during the crisis period (Sánchez-Sellero et al., 2016). Due to the temporary lapse since then, one could reasonably consider it of interest to continue this research focusing on the 2018 economic recovery, when employees' values and expectations may have changed. Although one recent study explores this phenomenon (Costello & Westerover, 2016), its results focus on the Millennial generation and do not contemplate a representative sample of the entire labour force.
This article provides two important contributions to the research on job satisfaction. First, we find that, despite salary cuts over recent years, pay satisfaction does not show a significant influence on job satisfaction. Our second contribution is the finding that a strong, direct and positive influence exists between the studied factors, namely supervisor support, work-life balance, career development, and job satisfaction.
These results can help firms and organizations understand the important influence of TC on employee satisfaction, thus allowing them to promote consistent human resource management policies that fulfil employee expectations.

Total compensation and its effects on job satisfaction
An analysis of the published papers in the SCOPUS database on the topics "Total Compensation" (TC) or "Total Reward" (TR), in comparison to papers on "Job Satisfaction" (JS), shows that TC+TR are less than 3% of the research carried out on JS (447 documents on TC+TR over 16.522 documents on JS). Research on JS began in 1939 (Williamson, 1939), and significantly increased in the 1970s. However, research on TC+TR is fairly recent, with the first published paper dating back to 1972 (Sullivan, 1972) introducing the concept of TC and developing the notion of "Indirect Compensation," which involves jobs with high levels of responsibility and professional development. It was not until the beginning of the 21st century that one could find extended literature on TC, but still far from the research on JS that is much more studied.

Figure 1
Published documents on "TC" or "TR", and "JS" on SCOPUS data base on 27 April 2021. Including documents classified as " Business, management & Accounting", "Economics, econometrics & finance" and "Social science" These results of published research on total compensation systems contrast with the research on compensation systems and their relationships with employee performance. Scientific interest in performance-based compensation systems dates back to Taylor's studies at the beginning of the 20th century (Dulebohn & Werling, 2007). In the period between wars, job evaluation systems emerged and were established to identify the value of each position to apply more equitable rewarding policies (Gooderham et al., 2018). Therefore, there has been remarkable research on the influence of incentive systems on productivity (Chiang & Birtch, 2010). However, it is still necessary to extend research on TC as a concept that goes far beyond tangible rewards as well as explore its impact on JS.
Compensation systems have been evolving in recent years. They were once based only on economic compensation. Now they have become enriched by intangible factors and benefits, such as the selection, training, and strengthening of competencies among employees who are considered suitable and have a greater ability to contribute to the company achieving its global objectives (Dinkin, 2015). The result of this evolution has come to be known as TC, a concept that encompasses not only all those elements related to wages and social benefits but aspects of the work context such as internal relationships, communication, transparency, supervisor support, professional perspectives, work-life balance, and even the company's values (Bessette, 2014). These elements of work context are becoming increasingly important for junior employees who tend to value more the company they are working for as well as some conditions concerning the work environment (such as transparency) than they do the position itself (Stewart et al., 2017).
TC has generated significant interest among researchers in the field of HRM (Hinote & Sundvall, 2015), with different approaches taken and investigators exploring the relationships between some of its elements and how these affect motivation and job satisfaction (JS). However, we remain far from reaching a widespread consensus, as made evident by some studies highlighting the misunderstanding between compensation and employee payroll (Marcos, 2011).
Research on attracting talent and generating loyalty and commitment (Cahill & Sedrak, 2012) has found that employees should experience a rich and satisfactory "trajectory" during their time with the company. This concept is commonly referred to as "Employee Value Proposition" (Veldsman & Pauw, 2018) and includes all aspects of TC.
The most commonly studied factors in the compensation research literature are fixed wages, variable salaries, benefits (Artz, 2010), flexible compensation (Artz & Heywood, 2015), bonuses, incentives, and stock options (Holland and Elder, 2006), restricted stock (Wu, 2011), employee stock purchase plans (ESPP), paid expenses, health programs, and pension supplements. Igalens and Roussel (1999) analysed other aspects besides flexible compensation, such as equity, benefits, and salary increase. Similarly, Shore et al. (2006) showed that internal equity is the most important predictor of motivation, while external equity is the most important predictor of turnover, job satisfaction, and employee performance (Koç, 2011). These studies call for further research on non-economic elements that employees perceive as highly-valued and could lead to higher satisfaction levels in the workplace.
The general goal of this research is to establish a causal relationship between TC and JC. In this sense, our general hypothesis is based on the idea that one's perception of satisfaction in the workplace is a consequence of several TC factors (Hinote & Sundvall, 2015).
Other more recent studies (Elrehail et al., 2019) have tried to determine the relationship between human resources management in the company and competitiveness. They analyse some classic elements such as resource planning, job analysis, and design or selection systems, but also other factors taking part in TC as compensation and rewards, training and development, or performance evaluation systems. In our study, we contemplate compensation from the perspective of pay Satisfaction and training and development from the perspective of career development.

Pay satisfaction and job satisfaction
The varied and many aspects of salary are among the most studied elements in the compensation research literature. Pay satisfaction has historically been a key factor in talent loyalty and retention in organizations (Stringer et al., 2011). However, Judge et al. (2010) research show a weak relationship between salary level and job satisfaction.
Investigating the effects of pay satisfaction on job satisfaction is important because, in economic terms, increasing wages for talent retention and loyalty is quite possibly the least sustainable approach. Furthermore, even though salary increases or extra-wage monetary elements are desirable and valued, it has a rather hygienic effect (Lambert et al., 1980). About twenty years ago, Igalens and Roussel (1999) analysed the relationships between the remuneration package structure, motivation, and job satisfaction. They showed that not all elements of economic compensation had the expected effect when applying a certain retributive logic that finally proved misleading. According to this study, individualized compensation for "exempt" workers (employees not subject to a specific schedule) could be a motivating factor under certain conditions. However, when considering flexible compensation for "non-exempt" workers (employees subject to schedules and, in many cases, to face-to-face marking), neither job satisfaction nor motivation improve. They also showed that benefits (such as expenses paid, provision of different types of services, or health plans) did not increase the motivation or satisfaction of either group. At that time, this result was surprising due to a general belief that any remuneration concepts should positively influence on motivation and job satisfaction. Moreover, the same study concluded that some elements influence satisfaction but not motivation when one might think that a motivated employee is very likely to be satisfied.
The study by Stringer et al. (2011), conducted on a group of grassroots employees, concluded that there was a strong relationship between pay satisfaction and job satisfaction. Similarly, Pouliakas and Theodossiou (2010) found that employees with lower salaries were less satisfied with their work-except in Nordic countries, where a low salary did not imply low quality of employment.
Other researchers (Chiang & Birtch, 2010) have focused on pay-for-performance (PFP) and its influence on obtaining more positive attitudes from employees, concluding that each person's level of identification with the organization's values constitutes a mediating factor between both variables. They also revealed that employees could see the PFP as an investment on the part of the organization, which in turn improves attitudes in the workplace. However, some previous studies found that the predicted job satisfaction of workers receiving performancerelated pay is lower on average than those in other pay schemes (McCausland et al., 2005).
Pay satisfaction has also been studied regarding the effects of adequate or inadequate equity (Shore et al., 2006). The study aimed to know the relationship between pay satisfaction and the perception of equity, with the conclusions being, on the one hand, that internal equity is the main predictor of motivation while, on the other, organizational support and external equity are predictors of job rotation and intentions to quit the organization.
According to Xavier (2014), reward systems are not effective in attracting and retaining employees, nor are they helpful in understanding and achieving organization's strategic objectives. This author also concludes that employees are generally dissatisfied with how incentives are managed and with the information they receive from the remuneration system and criteria for determining wage increases. The study concludes that HR management should pay more attention to changes in the environment (aging population and economic climate) and that further study of employees' salary expectations are necessary.
Based on these studies' findings, it seems necessary to test the hypothesis on the relationships between pay satisfaction and job satisfaction. Thus, Hypothesis H1 is stated as follows: H1 -Pay satisfaction (PS) has a direct effect on job satisfaction (JS).

Supervisor support and job satisfaction
One's relationship with their supervisor is one of the most frequent causes of quitting a job (Maertz et al., 2007) or adopting deviant behaviours (Kalemci et al., 2019). In contrast, when perceived positively, the supervisor relationship constitutes an important factor influencing loyalty (Shore et al., 2006). Research on the effect of supervisor support demonstrates a strong relationship between management behaviour and job satisfaction, productivity, and performance (Koç, 2011). Some research analyses the influence of supervisor support and skills development programs on perception and satisfaction, considering the importance of supervisors having social and relational skills (Bravo et al., 2017).
Other studies (Costello & Westover, 2016;Kaifi et al., 2012) focus on leadership factors that increase job satisfaction among employees and middle managers. They do so while considering generational differences: Baby Boomers, Generation X, Y, etc.
According to Eisenberger et al. (2002), supervisors contribute reduce turnover and increase talent retention. Azanza et al. (2013) point out that the effect of supervisor support on job satisfaction is important as long as the leadership is authentic. These results suggest Hypothesis H2, which is stated as follows: H2 -Supervisor support (SS) has a direct effect on job satisfaction (JS).

Career development and job satisfaction
Studies such as those by Chartrand and Camp (1991) and Baruch (2006) investigated the evolution of the main trends in career development, specifically from the perspective of professional success. Ismail et al. (2014) researched employees' perceptions regarding the effect of organizational support on professional development. They demonstrated a strong relationship between perceived career development support and job satisfaction. López-Andreu and Verd (2013) analysed how company policies and strategies affect career development opportunities, finding that career development was influenced more by individuals' characteristics (age, gender, or educational credentials) than by the resources that companies provided. Tobing's (2016) research finds results that partially contradict Ismail et al.'s (2014) study, demonstrating that career development does not significantly impact job satisfaction.
There may be variations between generations, as shown by Benson et al. (2018), who concluded that Generation X is more likely than Baby Boomers to experience higher satisfaction when they perceive development possibilities at higher levels of organizational commitment and greater compliance at work. These results suggest Hypothesis H3, which is stated as follows: H3 -Career development expectations have a direct effect on job satisfaction.

Work-life balance and job satisfaction
Work-life balance is a factor that became relevant during the 2008 economic crisis because many organizations adopted the tactics of shortening work hours and applying flexible schedules (Selva & Tresserra, 2013) as compensation in a scenario where wages were declining (or frozen, at best). Flexible work hours have become a central aspiration for new generations of workers who, as they clearly state, work to live but do not live to work (Kaifi et al., 2012). Balancing work and personal life is also very influential in employee behaviour, especially among "bluecollar" workers, thus generating a culture of improvement in their organizations (Lambert, 2000).
Previous research (Judge et al., 2005) shows a strong relationship between work-life balance and job satisfaction. Work-life balance could be a more relevant factor for employees with young children. However, Haar (2013) study shows that employees who can better manage their work-life balance obtain better results, regardless of their family context. In this same line, Brough et al. (2014) research showed that work-life balance is negatively related to rotation (Dousin et al., 2021) and stress while positively influences job satisfaction and family satisfaction. In the same way, Cheng et al. (2019) research found that work-life interference mediates the relationship between perceived job quality and intentions to stay.
Other studies suggest that a supportive leadership style and a mentoring and training programme may reduce work-family conflict and increase job satisfaction (Ru Hsu, 2011). However, previous research has shown that this relationship between flexibility and job satisfaction depends on employee characteristics (Origo & Pagani, 2008) or gender (Rasmussen et al., 2021).
Taking these previous studies as a reference, we state the following hypothesis: H4 -The work-life balance has a direct effect on job satisfaction.

Method
The study design is non-experimental, crosssectional, and confirmatory. The research follows a hypothetical-deductive procedure to test the set of hypotheses stated above.

Participants
This research involved 246 employees, including lower-level workers, technicians, and middle managers in Spain's manufacturing and service industries. All of them worked mainly in medium-and large-sized private sector companies. This study complies with all the ethical principles that regulate research in psychology and management. Completing the survey was voluntary and anonymous and offered no compensation for participating in this study. Table 1 shows the sociodemographic profile of this sample. Most of the participants have worked more than 5 years in their current organization, have a university education level, live mainly as a couple, and have at least one child. Sample profile n = 246 The sample group is mainly "nonexempt" (employees subject to schedule) (Bessette, 2014), and it is representative of the Spanish workforce, according to INE data.
Data was collected using a survey with 22 items. Each item had five possible values on a Likert scale of 5 points (1 = totally disagree, 5 = totally agree). Data collection was carried out in collaboration with Spain's primary trade union, Comisiones Obreras (CC.OO.). The survey was distributed via CC.OO. to union affiliates in the industrial sector in Catalonia. A sample of 4.000 individuals received the survey by email during the first half of 2018. The response rate was 6.2%.

Instruments
All items included in the questionnaire were part of measurement scales validated in previous studies. However, reliability and validity tests were also carried out in this study. First, every scale performed reliability tests using Cronbach's alpha coefficient. Subsequently, validity tests were performed using exploratory factor analyses with the maximum likelihood criterion to verify the convergence and unidimensionality of scale items. Bartlett's test of sphericity and the Káiser-Meyer-Olkin (KMO) measure of sampling adequacy were used to determine the suitability of the sample for performing factor analysis (Ledesma et al., 2002).

Pay Satisfaction. PS.
Pay satisfaction was measured according to 3 items on the Spector (1985) scale. Specifically, items were: "PS1. I feel unappreciated by the organization when I think about what they pay me"; "PS2. There are few rewards for those who work here"; and "PS3. I don't feel my efforts are rewarded the way they should be". A reliability test was performed using Cronbach's alpha coefficient. The pay satisfaction scale's alpha was 0.862, and Bartlett's test of sphericity had a significance level of 0.000 with a KMO of 0.723. A validity test was performed using exploratory factor analysis, demonstrating a high level of convergence, confirming a onedimensional scale with a single factor that explained 78.7% of the variation.

Supervisor Support. SS.
Supervisor support was measured using 8 items of the Lambert (2000) scale. Items are: "SS1. My supervisor is concerned about me as a person"; "SS2. Is helpful to me when I have a family or personal emergency"; "SS3. Feels each of us is important as an individual"; "SS4. Is helpful to me when I have a routine family or personal matter to attend to"; "SS5. Is concerned about the way the workers think and feel about things"; "SS6. Is understanding when I have personal or family problems which interfere with my work"; "SS7. Appears to know a lot about company policies that help employees manage their family responsibilities"; and "SS8. Keeps the things we tell him/her confidential". A reliability test was performed using Cronbach's alpha coefficient. The supervisor support scale's alpha was 0.943; KMO was 0.926, and the significance level was 0.000. Factor analysis showed that a single factor explained 71.7% of the variation.

Career Development. CD.
Career development perspectives were measured by three items from Xavier (2014) scale.
Specifically, selected items were: "CD1. I am satisfied with training and development opportunities in my organization"; "CD2. I am satisfied with the possibilities of developing a career in my organization"; and "CD3. I am satisfied with the climate for personal development in the organization". A reliability test was performed using Cronbach's alpha coefficient. The career development scale's alpha was 0.894; KMO was 0.746, and the significance level was 0.000. Factor analysis confirmed a one-dimensional scale, with a single factor that explained 82.5% of the variation.

Work-life balance. WL.
Work-life balance was measured by Haar (2013) scale. The items were "WL1. Nowadays, I seem to enjoy every part of my life equally well"; "WL2. I'm satisfied with my WLB, enjoying both roles"; and "WL3. I manage to balance the demands of my work and personal/family life well". A reliability test was performed using Cronbach's alpha coefficient. The work-life balance scale's alpha was 0.871; KMO was 0.714, and the significance level was 0.000. Factor analysis showed that a single factor explained 79.5% of the variation.

Job satisfaction (Dependent variable)
Finally, job satisfaction was measured using Judge et al. (2005) scale. Items were: "JS1. Most days I am enthusiastic about my work"; "JS2. I feel fairly satisfied with my present job"; "JS3. Each day at work seems like it will never end" (inverted); "JS4. I find real enjoyment in my work"; and "JS5. I consider my job rather unpleasant" (inverted). A reliability test was performed using Cronbach's alpha coefficient. The job satisfaction scale's alpha was 0.872; KMO was 0.812; the significance level was 0.000, and a single factor explained 66.6% of the variation. These results confirm that the measurement scales were reliable and valid.

Procedures and data analysis
Having validated scales and items, we built a measurement model to perform a confirmatory factorial analysis in AMOS v.22. The measurement model included control variables: age, sex, couple, and company size. GFI, CFI, and RMSEA indices were used to measure model fit (Hair et al., 2010). Confirmatory factor analysis results showed a good fit to the data: GFI = 0.891, CFI = 0.962 and RMSEA = 0.056. Table 2 shows the correlation matrix of latent variables.  Table 3. Two multi-group SEM analyses were performed to analyse potential differences between groups of employees. The first multigroup SEM analysis split the sample into two groups by sex. In the second multi-group analysis, the sample was split into two groups according to whether or not the participants' age was greater than 40 years (41 or more).
Before carrying out the multi-group SEM analyses, a measurement invariance test was performed for each pair of models (restricted and non-restricted). This showed that the groups were invariant in terms of the measuring instrument, thus allowing for multi-group analyses (Table 4). Multi-group SEM analyses were performed on AMOS v.22. At every step, one of the studied relationships was restricted and compared between groups (men vs. women, participants older than 40 years vs. 40 years or younger). A chi-square test was performed to assess the differences for each path. Table 5 shows the results of the multi-group SEM analysis by sex, and Table 6 shows the results of the multigroup SEM analysis by age. Table 5 Multi-group SEM analysis by sex *Men and women are not different at the path level. **Men and women are different at the path level, with a significance level greater than 90%. Table 6 Multi-group SEM analysis by age *Participants older than 40 years are not different from younger participants at the path level **Groups are different at the path level, with a significance level greater than 99%.

Results
The results confirm Hypothesis 2, according to which supervisor support is directly and positively related to job satisfaction, with a standardized regression coefficient of 0.329 and a very high significance level (p < 0.001). Likewise, Hypothesis 3 is confirmed, by which professional career development is directly and positively related to job satisfaction, with a standardized correlation coefficient of 0.273 and a very high significance level (p < 0.001).
Hypothesis 4 is also confirmed: work-life balance is directly and positively related to job satisfaction, with a standardized correlation level of 0.385 and a very high significance level (p < 0.001).
However, the results do not confirm Hypothesis 1 since the relationship between pay satisfaction and job satisfaction is not significant (p = 0.195). This hypothesis is only confirmed for younger employees (40 years or less). None of the control variables were significant. Finally, as the sq-multiple correlation is 34.4%, the studied model moderately predicts job satisfaction.

Discussion
This paper aimed to examine job satisfaction's effects from the total compensation factors: work-life balance, supervisor support, career development, and pay satisfaction. This research assumed that supervisor support, career development, work-life balance, and pay satisfaction are TC factors that may influence employees' job satisfaction.
The results show that the most influential is work-life balance. This finding is of great interest since previous research (Haar, 2013;Haar et al., 2014) showed that work-life balance also positively affects life satisfaction and mental health. Thus, this current study contributes to understanding how employees enjoy better psychological outcomes from managing their work and life roles. In addition, according to Brough et al. (2014), WLB is inversely related to stress and turnover, making WLB a key factor in TC management.
According to the results of the present study, the second most influential factor is supervisor support, which aligns with some previous research (Maertz et al., 2007;Costello & Westover, 2016). Existing research confirmed that good perception regarding supervision is critical in reducing employee turnover as well as in improving productivity and performance. This research confirms that supervisor support is a relevant factor in TC and, furthermore, it has a positively influences on job satisfaction.
Career development is the third factor with a direct effect on job satisfaction, although to a lesser extent than the previous two factors. Previous studies had provided contradictory results on this factor (Ismail et al., 2014;Tobing, 2016). While the study by Ismail et al. (2014) showed a strong relationship between job satisfaction and career development, the study by Tobing (2016) concluded that this relationship was not relevant. Our study aligns with Ismail et al. (2014) findings by showing that, for Spanish employees, career development has a positive impact on their job satisfaction. This implies a challenge for HR departments, which should be able to combine business growth with professional development.
Finally, even though pay satisfaction has been hypothesized to be a predictor of job satisfaction, it has not proven significant in the general sample. This is in line with other studies (Judge et al., 2010) that reached the same conclusions.
These results are equally confirmed for the different studied groups, except for two instances. On the one hand, although pay satisfaction does not significantly influence job satisfaction in the general sample, it is a significant factor for the younger group (40 years old employees or less). Thus, in other words, pay satisfaction is directly related to job satisfaction for younger employees. It would be reasonable to conclude that younger generations, in general, are paid less than older generations of employees. When we consider that the pay satisfaction factor is shown to be influential on job satisfaction for younger employees, it confirms that this group (usually paid less) values their salary more so than older employees (who are generally better paid). On the other hand, although both men and women perceive work-life balance to be an influential factor in job satisfaction, this factor is more relevant for men than women.

Conclusions
This study confirms that some TC factors impact JS using structural equation models on a sample of 246 employees working in Spanish companies.
Our study results show that work-life balance, supervisor support, and career development are the most influential factors.
According to the study results, these three factors explained 34.4% of employees' job satisfaction. It is necessary, therefore, that managers conceive these factors to be an integral part of TC and that, consequently, promote policies of work flexibility and career development plans while encouraging supervisors to support their employees. These fundamental elements of TC will lead to employees experiencing a more satisfying trajectory through their organizations.
It is important to broaden the notion of compensation by incorporating the intangible factors that we analysed in this study beyond the old paradigm of fixed and variable economic rewards, which is still deeply rooted in the managers' minds of many companies. To this end, it is necessary to extend research on TC to increase employees' satisfaction and commitment to the company's projects.
This study on the influence of TC on Job Satisfaction is limited to the four factors studied. However, we believe other factors not reviewed here, such as company values, transparency, or communication, can also be relevant for future research.