The effect of psychological contract on job related outcomes: The moderating effect of stigma consciousness

Abstract This study examines how employees’ tendencies with regard to relational or transactional contract could influence job related outcomes. The typology then relates to the stigma consciousness that each employee has regarding their employment status, as either temporary or permanent workers. This study argues that each worker has their own expectations with regard to being stigmatized due to their status employment, and this could then strengthen/weaken the relationships between relational or transactional contract on job related outcomes. The proposed hypotheses are validated through two cross-sectional studies in two different organizations in Indonesia. The findings indicate that there are positive relationships between individuals who predominantly hold relational contract and their job satisfaction and job performance, and less intention to quit. However, the opposite effects for transactional contract are not revealed in this study, which might be due to the cultural values of the sample. Further, the positive relationships of relational contract on job satisfaction and contextual performance are weakened when the respondents have high stigma consciousness, while the relationship with intention to quit is strengthened. Additional findings indicate that individuals who predominantly hold transactional contract tend to have less job satisfaction when they are highly stigma conscious. The academic and managerial implications of this work are also discussed in this study.


PUBLIC INTEREST STATEMENT
This study investigates the relationship between psychological contract and job related outcomes as well as for whom the relationship between two is stronger/weaker. Based on surveys in two different organizations, the results indicate that respondents who predominantly hold relational contract tend to have less intention to quit and are more satisfied with their current job, while also having higher task and contextual performances. Even though respondents who predominantly hold transactional contract saw the opposite effects to those who hold relational contract, the differences are not significant, and this may be due to the cultural context in which the study was carried out. These findings indicate that managing employees' psychological contract and reducing the negative effects of stigma consciousness are important with regard to improving job related outcomes.

Introduction
Psychological contract has received considerable attention from both academics and practitioners, with the Social Sciences Citation Index (SSCI) noting that the number of studies published on this issue between 2017 and early 2021 is almost three-hundred times the number that was published in the first 30 years since the concept was proposed by Rousseau (1989). This rise in interest is mostly due to the changing landscape with regard to employee perceptions of and reactions to the employment relationship, related to such factors as outsourcing, increased reliance on temporary workers, and demographic diversity, among others (e.g., Boyce et al., 2007;Coyle-Shapiro et al., 2019). Limited studies empirically test how the content of psychological contract, either relational or transactional contract, on organizational outcomes (Garcia et al., 2021;Raja et al., 2004).
Despite the direct consequences of psychological contract among employees, few studies discuss for whom the content of a contract (relational or transactional) magnifies the relationship with job related outcomes. In the work environment, every person expects to be respected by others (Pinel & Paulin, 2005); but some individuals might expect higher or lower levels of respect (Wells et al., 2021). Individuals who are targets of stigma will have lower expectations, as they will regularly experience insinuations that they do not merit respect (e.g., Crocker & Major, 1989;Woodhall-Melnik, 2018). Following Pinel (1999), this study defines stigma consciousness as the expectation that one will experience prejudice and discrimination, or be negatively stereotyped by others (Pinel, 1999) because of their attributes in the work environment. However, not all individuals with alienated status have the same expectations about the degree to which others will discriminate against them (Cortina et al., 2017). In other words, individuals who are high in stigma consciousness are more likely to believe and be worried that discrimination is directed at their marginal group in general, and at them personally (Shigihara, 2018;Wells et al., 2021).
Stigmatization occurs when individuals are being treated in a devalued manner because they possess certain attributes that are not valued in a particular context Woodhall-Melnik, 2018). In the work environment, the classic issue of stigmatization is the different status of permanent and temporary workers, in which the latter are particularly attentive to prejudice cues, although they may be being subliminal (Kaiser et al., 2006). Boyce et al. (2007) argued that there is stigmatization directed at temporary workers from their permanent colleagues and the organization as a whole, because discrimination based on job status is inherent in many organizational policies and practices. Although many national laws and regulations require equal treatments between permanent and temporary workers (Scheel et al., 2013), in common practice many that organizations distribute benefits differently between the two which will engender a perceived incongruence (Shanock et al., 2010). Temporary workers usually earn less money than their counterparts and are less likely to have health and pension benefits (Kalleberg, Reskin, and Hudson, 2000), are commonly employed in jobs with low complexity (Parker, 1994), and are less likely to be listened to when proposing any adjustments to organizational practices (Aronsson, 1999;Bernhard-Oettel & Isaksson, 2005).
On the other hand, Von Hippel and Kalokerinos (2012) and Banerjee et al. (2012) showed that permanent workers may feel threatened by the existence of temporary workers, and such feelings are exacerbated when permanent employees believe that temporary workers desire to take their jobs (e.g., Davis-Blake et al., 2003;Stephan and Stephan, 2000). In other words, employment status could be an attribute that might engender stigmatization in the workplace, in which some workers might feel undervalued by the management (mostly temporary workers) while some workers might feel threatened by the existence of other types of workers (mostly permanent workers). This notion is in accordance with the finding of Boyce et al. (2005) that the level of stigmatization among temporary workers varies, which itself is supported by the idea in that the level of stigmatization of person feels depends on the extent of their stigma consciousness (Pinel, 1999). Based on these earlier works, this study proposes that the relationship between an individual's psychological contract which represents their beliefs regarding the conditions of the exchange relationship (Rousseau, 1989) and their job related outcomes is weakened if they are more stigma conscious.
In addition, Fiske (1991) posited that culture is a primary component in the choices people make as to how exchanges occur. To date, the concept of psychological contract has been well supported in most studies in Western (e.g., Coyle-Shapiro & Conway, 2005;Zhao et al., 2007) and Eastern contexts, with the latter dominated by works on China (e.g., Hui et al., 2004;Z-X. Chen et al., 2008), but the theory still needs to be tested in different settings (Baruch & Rousseau, 2019;Gelfand et al., 2008). This study was thus conducted in Indonesia. Even though both Indonesia and China are in East Asia, and are expected to have cultural overlaps, significant differences between the two exist, especially with regard to ethnic and religious diversity, as well as economic growth (Kutaula et al., 2020;Mahmood et al., 2019). This study should thus enrich the psychological contract literature.
In summary, this study differs from previous psychological contract research in two ways. First, it discusses for whom relational and transactional contracts magnify the relationships with job related outcomes. This study proposes a moderating variable to examine this question-stigma consciousness. The authors expect that a consideration of these variables can contribute to the existing literature on psychological contract and stigma consciousness, as well as provide advice for practitioners. Second, the selected research sites are unique-private and public organizations in Indonesia-compared with previous studies, and presenting the two settings together is expected to increase the generalizability of the proposed hypotheses and findings.

Psychological contract
Following most of the existing literature, this study defines psychological contract as an individual's subjective understanding of promissory-based reciprocal exchanges with their employer (Rousseau, 1989), which is shaped by the organization within which such exchanges occur (Rousseau, 1995). This subjective aspect refers to individuals' beliefs in the existence of an exchange agreement (Millward & Brewerton, 1999), while the reciprocal aspect refers to beliefs regarding the mutual obligations of both parties to the relationship (McLean Parks et al., 1998;Rousseau, 1990).
There are two major research streams for psychological contract. The first emphasizes how the content of psychological contract influences key organizational outcomes. For example, individuals who hold a predominantly relational contract tend to have more job satisfaction and affective commitment, and therefore less intention to leave the organization (e.g., Garcia et al., 2021;Raja et al., 2004). Hui et al. (2004) reported that relational contract have a strong relationship with extra-role behaviors among Chinese employees, while the relationship with transactional contract is weaker. The second stream of studies focuses on whether employers fulfill or breach their obligations or promises to employees (Deng et al., 2018;Wiechers et al., 2019). For example, prior studies indicated that when a breach occurs, employees tend to perform more poorly, engage in greater job search activities (e.g., Raja et al., 2004;Robinson, 1996), and cut back on job performance (e.g., Robinson & Morrison, 1995;Zhao et al., 2007). On the other hand, when individuals believe that their employers fulfill their obligations, their job performance improves (e.g., Turnley et al., 2003). This study focuses on the first stream of research, which has been less investigated by previous studies (e.g., Garcia et al., 2021;Hui et al., 2004).
According to Rousseau (1995Rousseau ( , 2000, psychological contracts vary along both time frame and performance requirement dimensions, yielding four different forms: transactional, relational, balanced, and transitional contracts. This study uses the first two dimensions of psychological contract, since the concept itself is clearly defined and widely used in most of the related empirical studies (e.g., Hui et al., 2004;Raja et al., 2004;Turnley et al., 2003). Transactional contracts refer to a short-term exchange of specific benefits and contributions that are highly monetary or economic in nature (Hui et al., 2004), and which take place over a finite period (Robinson et al., 1994). In contrast, relational contracts are broader and more long-term oriented, as they are not restricted to purely economic exchanges but also include socio-emotional obligations (Morrison & Robinson, 1997;Rousseau & McLean, 1993). This study regards these two dimensions of psychological contract as coexisting, and thus an individual can to some extent have both simultaneously and mutually exclusive relationship (Rousseau, 1990). In addition, this study operationalizes psychological contract as an employee's obligations, based on the framework in Guest (2004).
Theory provides good grounds for expecting that relational contract would facilitate positive personal outcomes more than transactional contract would (Rousseau, 1995;Rousseau & McLean, 1993). When individuals predominantly hold relational contract, the resources exchanged with the organization span from concrete (economic resources) to abstract ones (social-socioemotional resources) (Foa & Foa, 1974, 1980. A person's job satisfaction thus does not merely rely on the financial payoff they receive, and many other organization factors can affect it (Cavanaugh & Noe, 1999;Raja et al., 2004). In other words, if individuals are dissatisfied with one aspect of the exchanged resources, say the salary, they will consider many other aspects, such as their friendships with colleagues or a pleasant working environment, when determining their job satisfaction. Consequently, relational contract oriented individuals tend to have greater job satisfaction. In contrast, individuals with a predominantly transactional contract tend to have less job satisfaction, as it is more narrowly based on the direct monetary benefits they accrue from their work. Over time, the value of the monetary benefits they receive deteriorates psychologically (when colleagues get promoted, or as they become accustomed to their level of income) or in real terms (due to inflation). As a result, individuals with a predominantly transactional contract orientation have less job satisfaction compared to individuals with a predominantly relational contract (Raja et al., 2004). Moreover, if transactional contract individuals believe that their employer has no fulfilled their promise-breach, this will lead to an immediate reduction in job satisfaction ((Deng et al., 2018;Wiechers et al., 2019;Zhao et al., 2007), in contrast to the reaction of relational contract individuals, who consider other rewards as part of their calculation of how their satisfied they are at work.

Stigma consciousness
Although person expects to be respected by others (Pinel & Paulin, 2005), some individuals might expect higher or lower levels of respect. Individuals who are targets of stigma will have lower expectations, as they will regularly experience insinuations that they do not merit respect (e.g., Crocker & Major, 1989;Gaertner & Dovidio, 1986). Following Pinel (1999), this study defines stigma consciousness as the expectation that individuals will experience prejudice and discrimination, or be negatively stereotyped by others (Pinel, 1999) because of their attributes in their work environment. However, not all individuals with alienated status have the same expectations about the degree to which others will discriminate against them. In other words, individuals who are high in stigma consciousness are more likely to believe and be worried that discrimination is directed at their marginal group in general and at them personally.
Stigmatization occurs when individuals are being treated in a devalued manner because they posses certain attributes that are not valued in a particular context . In the work environment, the classic issue of stigmatization is the different status of permanent and temporary workers, in which the latter are particularly attentive to prejudice cues, although these maybe subliminal (Kaiser et al., 2006). Boyce et al. (2007) argued that there is stigmatization directed at temporary workers from their organizational and permanent colleagues, because discrimination based on job status is inherent in many organizational policies and practices. As noted by De Cuyper et al. (2008), temporary workers earn less and have less access to fringe benefits (OECD, 2002), and are less likely to be heard when proposing any changes to organizational practices (Aronsson, 1999;Bernhard-Oettel & Isaksson, 2005). These conditions make temporary workers prone to be stigmatized, even though this does not necessarily mean that an individual will automatically be stigmatized in a given situation . Moreover, if individuals have a higher level of stigma consciousness, they will have a greater expectation that they will be treated unfairly, regardless of their job status, because of their lower self-esteem .

Hypotheses development
Previous studies reported that relational contract relates positively to expected job tenure, while transactional contract relates negatively to it (Millward & Hopkins, 1998;Rousseau, 1990). When individuals predominantly hold relational contract, they perceive and expect that their employment relationships are open-ended and broad, not merely based on economic exchanges but also social ones, and this engenders loyalty toward the organization (Cavanaugh & Noe, 1999;Morrison & Robinson, 1997;Rousseau & McLean, 1993). On the other hand, when individuals' beliefs are dominated by transactional contract, they tend to have short-term employment relationships and mainly focus on economic exchanges (Blau, 1964), and thus are more likely to quit when other organizations offer greater monetary benefits. These arguments are also supported by the finding of Raja et al. (2004) that individuals who predominantly hold transactional contract have greater intention to quit than those with relational contract.
Social exchange theory posits that social relationships basically consist of exchanges of both economic and social resources (Blau, 1964). When individuals receive their partner's exchanged resources, they feel indebted and obliged to reciprocate (Gouldner, 1960). According to psychological contract (Rousseau, 1989), these individuals are more likely to demonstrate reciprocal behavior toward employers that have fulfilled their promises. With a broad range of exchanged resources, these individuals put many factors (social-socioemotional and economic- Foa & Foa, 1974, 1980 into their obligation account. As a result, this reciprocation tends to be stronger and leads to behaviors that directly contribute to organizational performance (Borman & Motowidlo, 1997;Uddin et al., 2019). Moreover, since these individuals have long-term relationships with the organization, they have greater motivation to perform in-role behaviors as part of efforts to keep the organization sustainable. On the other hand, since transactional contract individuals are shortterm oriented (Rousseau, 2000), they tend to perform their job as fast as possible and care less about the quality of their performance. This is because they usually always calculate whether their fulfillment will be reciprocated on the employer side-under-obligation psychological contract (De Jong et al., 2009), and so tend to reduce their performance, leading the conditional of mutual low fulfillment. As reported by Zhao et al. (2007), transactional contract individuals tend to react immediately when the employer fails to deliver on their promises by reducing their in-role performance compare to employees with relational contract. Such actions are also seen as justified when these employees think that they may soon leave the organization (De Clercq et al., 2020;Raja et al., 2004).
In contrast, relational contract individuals tend to have greater contextual performance. As described by Borman and Motowidlo (1997), contextual performance includes those behaviors that contribute to the organization by fostering a positive social and psychological climate. Individuals with a relational contract orientation take more account of the social-socioemotional and economic features in their exchanged resources (Foa & Foa, 1974, 1980. When they receive social or emotional favors from their colleagues or employer, they will feel indebted and reciprocate to maintain the mutual obligations, and thus will have even greater contextual performance. Consequently, the organization will have a more positive social and psychological climate. In addition, Farh et al. (2004) reported that in China, as in Asia in general, relationships have pivotal roles in the work place and individuals are motivated to behave in ways that strengthen them. However, by perceiving economic resources as the only exchanged resources, transactional contract individuals have less motivation to perform any extra-role behavior that will not affect their monetary benefits (Ahmad and Zafar, 2018;Hui et al., 2004). Consequently, they will perform less with regard to contextual performance. Therefore, this study hypothesizes that: H 1 : There will be a direct negative relationship between the extent to which individuals predominantly hold relational contract and their (a) intention to quit, and a direct positive relationship with their (a) job satisfaction, (b) task performance and (c) contextual performance.
H 2 : There will be a direct positive relationship between the extent to which individuals predominantly hold transactional contract and their (a) intention to quit, and a direct negative relationship with their (a) job satisfaction, (b) task performance and (c) contextual performance.
As outlined in an earlier section, research suggests that targets high in stigma consciousness are more likely to disengage-either psychologically or physically-from stigmatizing situations than other individuals , and this disengagement appears to have a self-protective function . Even though relational contract oriented individuals have an openended relationship orientation, a greater expectation that other parties will discriminate against them could increase their intention to quit. This is because these individuals will aim to reduce any possibility that these negative emotions are reproduced (Lawler & Thye, 1999), and the only way to do this may be to move another organization. A similar effect maybe even more significant for individuals who predominantly hold transactional contract, since their short-term relationship orientation is magnified by greater stigma consciousness, in which their intention to quit would be strengthened.
Moreover, individuals who predominantly hold relational contract subjectively perceive that their exchanged resources are broad-based (Morrison & Robinson, 1997;Rousseau & McLean, 1993). However, when these individuals expect that they will experience discrimination because of particular attributes they possess, the positive relationship between relational contract and job satisfaction tends to decrease (Handy et al., 2020). This is because when individuals perceive that they receive negative sentiments from their exchanged partners, the quality of social exchanges between them deteriorates (Golden & Veiga, 2018;Shelly, 1993), and thus their job satisfaction will decrease. Similarly, even though transactional contract oriented individuals have less regard for non-monetary aspects of their exchange process, perceiving that others will discriminate against them would still psychologically and socially influence their overall assessment of their jobs. As a result, their level of job satisfaction is also lower than if they have less expectation of being discriminated against by others.
The previous section described how a broad range of exchanged resources increase the possibility for relational contract oriented individuals to have more obligations to their organizations, which need to be reciprocated by performing better in their jobs. By performing better, such individuals could support the sustainability of their organizations, and thus facilitate their longterm relationship orientation. However, this will not be the case when individuals have greater stigma consciousness. Consistently expecting that others will discriminate against them substantially reduces the resources that such individuals are willing to exchange with their organizations (Shelly, 1993), and thus, their task performance will be lower (Handy et al., 2020). In a similar manner, individuals who predominantly hold transactional contract may reduce their task performance when they are more stigma conscious, since negative emotions lead individuals to reduce the resources they are willing to exchange (Lawler & Thye, 1999).
The negative moderating effect of stigma consciousness is even more significant with regard to contextual performance. Individuals with a relational contract orientation will be more sensitive when they expect that their exchange partner will discriminate against them. They will reduce their positive behaviors, including assisting coworkers when they need help and being courteous and respectful to peers (Kiazad et al., 2018;Van Scotter et al., 2000). Moreover, since the expectation to be discriminated against originates from their exchange partners, the motivation to maintain a positive social and psychological climate in their exchange relationship will be reduced. In line with this, transactional contract oriented individuals also reduce their contextual performance due to negative feelings with regard to their exchange process. Consequently, the following hypotheses are proposed: H 3 : Perceived stigma consciousness moderates the relationship between relational contract and job related outcomes in such a way that the negative relationship between relational contract and (a) intention to quit is weakened when the respondents have high stigma consciousness, as are the positive relationships between relational contract and (b) job satisfaction, (c) task performance, and (d) contextual performance.

Methods
Two studies are conducted in this research, with the second based on the promising results of the first. Moreover, the two studies consider different types of organization (Study 1 is a private organization, while Study 2 is a public organization), which is expected to increase the generalizability of the findings. Previous studies mainly examined only one type of organization, and McDermott et al. (2013) thus noted that simultaneously conducting studies in two different types of organization could offer interesting results, particularly for the psychological contract literature. Furthermore, since the first study used self-reported data from a single organization, the second study was planned to have a better research design by having multiple sources. By asking direct superiors to assess their subordinates' performance, the second study offers some insights that are rarely discussed in prior psychological contract studies (e.g., Hui et al., 2004;Raja et al., 2004). Our research model is shown in Figure 1.

The organization
The privately-owned organization is a joint venture company in the oil industry. The company is about 70% owned by the government of the People's Republic of China (PRC), and is the third largest oil company in China. It has 24,000 employees around the world, and their operations in Indonesia have 884 employees. The Indonesian government requires joint venture companies to hire local people in their operations. In order to maintain efficiency, these firms tend to hire both permanent and temporary employees, with a higher proportion of the latter in order to reduce the costs related to permanent staff, such as full coverage of health insurance and child-education support (Boyce et al., 2007). With regard to the temporary workers, all of them are hired through an agency (Connelly & Gallagher, 2004) to comply Indonesian laws. Their working contracts are extended on a yearly basis, although many workers remains employed for more than 20 years as temporary ones. Furthermore, although the sample is composed of engineers, production workers, and administration staff, the work carried out by them could be described as simple and routine, with low task variety and complexity. This is because these employees are working in an oil exploitation site, which presents far less variety than an exploration site. In terms of benefits, temporary workers receive less than permanent ones, as well as less employment stability. In addition, even though the company is mostly owned by a foreign firm, there is no discrimination among local and expatriate (Chinese) workers, and the share of jobs between the two groups is roughly proportional, in managerial as well as other positions, with more than 90% of the employees being Indonesian citizens.

The environment
This study was conducted at a time when the economy was functioning very well and unemployment was low. Specifically, employee turnover rates in Indonesian oil companies were relatively low, due to the fact that oil prices were high (above 100 USD per barrel), and firms needed more employees to aggressively explore and exploit sources of oil. The Indonesian economy also grew by 6.0 % for the duration of the study.

Procedures and sample
This study developed its measurements based on a review of the literature. In order to maximize functional and conceptual equivalence during the translation process, the questionnaire was written in English and then translated into Indonesian using a double-back translation method. Following the procedure of Harkness (2003), an initial pre-test was conducted with 24 respondents and revisions were made based on their feedback before distributing the final instrument to the target respondents. The qualified sample consists of employees of one firm owned and operated by foreign companies, and it was selected because almost 70% of its employees are hired under short-term contracts which need to be renewed annually. It is thus a suitable setting to test the hypotheses developed in this study. The survey was sent out with a university-headed cover letter from the researcher, in which respondents were asked to express their opinions about the questionnaire items for all the research constructs, and were promised complete anonymity.
Data were collected over a two-month period from the beginning of March, 2008 to the end of April, 2008. Before mailing the questionnaires, emails were sent to the related human resources managers to obtain permission for the survey and to identify suitable employees to participate in the study. In order to get more balanced perceptions, this study also asked the workers' union to identify suitable respondents. Eventually, a total of 300 questionnaires were mailed to respondents (150 recommended by the HR department, and 150 recommended by the union), who voluntarily participated in this work. This study only targeted employees at the production site and head office on Java Island. Out of 300 potential respondents, 165 actually completed the questionnaire. However, since there were some missing variables for two of the questionnaires, this study dropped these from further analysis, giving a total of 163 usable questionnaires, for a response rate of 54.33%. In order to test for sample selection bias, this study compares the respondents who were recommended by the HR department and those who were recommended by the union. A chi-square and Mann-Whitney tests were used to examine whether there was any systematic difference in terms of age, tenure, and status among the respondents. The results indicated that p values were greater than 0.050 for these three characteristics, and thus, the results are generalizable with regard to the sampling frame.

Survey design features
Due to constraints imposed by the organization, this study uses self-reporting to measure all of the variables reported in Study 1. As such, a number of steps are taken in the survey design to minimize the potential impact of common method variance (Podsakoff et al., 2003). For example, this study separated the predictor and criterion measures on the survey, and placed objective demographic questions in between (Lavelle et al., 2009). In addition, despite the anonymity of respondents, this study also inserted other measures that were not relevant to the purpose of the present research.

Stigma consciousness
Following the procedure of Pinel and Paulin (2005), this study adapts the items of stigmatization from the work of Pinel (1999) and rewords them for a work context. The original items consist of 10 questions, and this study selects items with factor loadings higher than 0.50, which is generally considered as significant (Hair et al., 2006), and is consistent with the two studies in Pinel (1999). However, the pre-test results (N = 24) indicate that item 7 has a factor loading of less than 0.4, and thus this item is discarded in the main study. The retained items are as follows: (1) "When interacting with colleagues, I feel like they interpret all my behaviors in terms of my job status (status of employment)," (2) "Most employees judge their colleagues on the basis of their job status (status of employment)," (3) "My job status (status of employment) influences how my colleagues act with me," and (4) "I seldom think about my job status (status of employment) when I interact with my colleagues" (reversed item). The reliability score was 0.94.

Job satisfaction
Consistent with this earlier definition, this study sees job satisfaction as an individual's positive attitude toward their job in general (Weiss & Cropanzano, 1996). This study measures job satisfaction with the overall measurement: "All things considered, I am satisfied with my current job." Prior research has shown that the validity of this one-item general job satisfaction measure is comparable to that of multi-faceted measures involving multiple items (Begley et al., 2006;X. P. Chen et al., 1998).

Intention to quit
Intention to quit refers to the subjective probability that an individual will leave his or her organization within a certain period of time (Zhao et al., 2007). This study measured it with a threeitem scale extracted from the Michigan Organizational Assessment Questionnaire (Cammann et al., 1983;X. P. Chen et al., 1998;Raja et al., 2004). Items included "I will probably look for a new job in the next year," with a Cronbach's α of 0.91. All the research variables in Study 1 are assessed by using a seven-point Likert scale.

Characteristics of respondents
The majority of the respondents are male (65 percent) and married (80 percent). One third are under 30 years old, and the biggest group is between 31 to 40 years old (35.60%), and the rest are over 41. Moreover, the majority of the respondents only graduated from senior high school (53.40%), with around 40% earning a university degree (diploma, bachelor, or master), and around 6% of respondents did not indicate their educational level. Sixty-five percent of the respondents are temporary workers, while 35% are permanently hired by the company, and this data has been double-checked with organizational records. Only 12% are administrative staff, 20% are engineers, and the rest have production roles (drilling, machine operators, and so on) in the company. There are 29 supervisors (18.40%), and the rest are workers (80.60%). Finally, 41% of the respondents have been in their current positions for less than three years, while more than 24% have been so for more than 10 years.

Study 2
Study 2 was designed to build on the promising results of Study 1 in several ways, but specifically, it provides some methodological enhancements over its predecessor. The biggest limitation of Study 1 was the use of one only source, the respondents, which raised the possibility of commonmethod bias (Podsakoff et al., 2003). Unlike Study 1, the measures of behavioral outcomes (job performance) in Study 2 are provided by the respondents' direct supervisors.

The organization
The organization is a top-tier public university, located in an urban area in the second largest city in Indonesia. The university has 13 colleges and 32 departments, with more than 22,000 full-time students, and nearly 2,000 graduate students. The university has 961 permanent administrative employees and 1,494 full-time faculty members, with almost the same number of temporary administrative and academic staff. In terms of benefits, temporary workers receive less than permanent ones, as well as less employment stability. The temporary workers in are directly hired by the university (Connelly & Gallagher, 2004) rather than through agency. According to Wilkin (2013), both have been regarded as temporary workers, but the differentiating factor is the employer, whether the agency or direct organization.

The environment
This study was conducted at a time when the Indonesian economy was being hurt by the ongoing global economic crisis. However, employee turnover rates in the education industry are almost zero, due to the fact that it is not as dynamic as other industries in Indonesia. During this period, Indonesia's economic growth was projected to be over 4%, while other countries in the region were expected to experience negative growth. Therefore, the general economic environment is only slightly different to that in Study 1.

Procedures and sample
The selected sample is the administrative employees of one public university in Indonesia. As described above, about half of the 2,000 administrative employees are full-time, while the other half are temporary, and it is thus a very suitable setting to test the hypotheses developed in this study. The survey material included a university-headed cover letter from the researcher, in which the respondents were asked to express their opinions about questionnaire items for all the research constructs, and were promised complete anonymity. Data were collected over a twomonth period, from the middle of February, 2009 to the middle of April, 2009. Before the questionnaires were sent out, the human resources director gave permission for the study and access to the employee database. In each college and work unit, the university has four sub-divisions: academic, studentships, finance, and general appliances. Based on the database, this study randomly chose two full-time and two temporary administrative employees from each subdivision to participate in the study. The leader of each sub-division was invited to assess their subordinates in terms of their job performance. Among the 208 employees and 52 leaders invited, more than 60% agreed to participate, giving final responses from 113 employees and 31 leaders, although only 105 responses were perfectly matched.

Survey design features
The questionnaires used in Study 2 have two parts. The first is targeted at administrative employees, and is used to self-report measurements for all the variables reported in Study 1. As before, this study separates the predictor and criterion measures on the survey, with objective demographic questions placed in between them (Lavelle et al., 2009). Other measures that are not relevant to the purposes of the present research are also included. The second part is targeted at the administrative leaders of each sub-division, and this is used to obtain the supervisors' assessments of their subordinates' job performance.

Measures
The measurements in Study 2 are similar to those in Study 1, and are assessed by using a sevenpoint Likert scale. The only difference is that the job performance is assessed by the supervisor, as described below:

Job performance
Researchers mostly agree that there are two dimensions to job performance, task and contextual performance (e.g., Motowidlo & Van Scotter, 1994;Van Scotter et al., 2000). Murphy (1989) defined task performance as the accomplishment of duties and responsibilities associated with a given job, and an example item is "He/she performs the tasks that are expected of him/her." This study measures it by using seven items developed by Williams and Anderson (1991), and their Cronbach's αwas 0.93. In contrast, contextual performance involves behavioral patterns that support the psychological and social context in which task activities are performed (Van Scotter et al., 2000). Contextual performance is measured by using interpersonal citizenship performance, which is based on the concept of interpersonal altruism (four items; Cronbach's α ¼ 0:92) in Coleman and Borman (2000). This closely reflects the definition of contextual performance proposed by Borman and Motowidlo (1993), and items included "He/she helps other organizational members."

Characteristics of respondents
The characteristics of the subordinates are as follows: The proportion of male and female respondents is almost equal, and a majority of the respondents are married (79%). Less than 15% of respondents are under 30 years old, and the biggest group is between 31 and 40 (44.50%), while the rest are over 41. Moreover, the majority of the respondents only graduated from senior high school (53.30%), with around 26% earning a university degree (diploma, bachelor's, or master's), and around 20% did not indicate their educational level. Thirty eight percent of the respondents are temporary workers, while 62% are permanent ones, and this data has been double-checked with organizational records. Seventy-seven percent of the respondents are administration staff, 5% are technicians, and the rest have other roles in the organization. There are 15 supervisors (13.50%), 61 at staff level (70.90%) and the rest did not indicate their positions. Finally, 46% of the respondents have been in their current positions for more than ten years, while more than 14% have been so for less than three years.
The majority of the supervisors are male (61%), married (9%), and over 40 years old (96%). Eleven percent of the respondents only graduated from senior high school, with the rest having a university degree (diploma, bachelors, or masters). More than 90% are classed as middle-level managers, while 7% are supervisors. All of them have permanent employment status. Finally, one third of the respondents have been in their current positions for less than three years, while more than 30% have been so for more than 10 years. Table 1 presents the characteristics of respondents from both studies:

Reliability and validity of measurement constructs
The construct validity is assessed using the guidelines of Anderson and Gerbing (1988). First, the exploratory factor analysis for all the items resulted in factor solutions, as expected theoretically. The Cronbach's α for each variable is greater than 0.700 in Study 1 and Study 2. Second, we used confirmatory factor analyses (CFA) to assess the convergent validity of the measures. All loadings exceed 0.600, and each indicator t-value exceeds 10 (p < .001), thus satisfying the CFA criteria (Hair et al., 2006). Most of research variables had eigen-values of over 2.000 and variances explained above 0.600. All these figures support the overall measurement quality given a particular sample and number of indicators (Gerbing and Anderson, 1992), and the measures thus demonstrate adequate construct validity and reliability. The results are presented in Table 2.
To assess the potential impact of common method bias in the present study, the discriminant validity is tested in three steps. First, a Harman one-factor test is conducted (Podsakoff & Organ, 1986) which loads all the variables into a principal component factor analysis. The results reveal that no single factor dominates: Study 1 generated five factors with 72.256% of the total variance, and factor 1 is only 19.888% of the variance; Study 2 generated 7 factors with 77.656% of the total variance, and factor 1 is only 17.007% of the variance. Second, the variance-extracted percentages for any two factors are compared with the square of the correlation estimate between them (Fornell & Larcker, 1981). Table 3 reports the inter-factor correlations and their squared values. Each of the variance-extracted estimates is greater than the corresponding inter-factor squared correlation estimates (i.e., has values above the diagonal). Finally, the χ 2 -difference test is performed for each pair of factors that have correlation values above 0.40 (11 cases) by using the common method factor (Conger et al., 2000). All cases result in a significant difference, which further indicates that the pairs are not collinear (Anderson & Gerbing, 1988). The discriminant validity among research constructs is thus further confirmed.
Correlations show that relational contract is positively related to job satisfaction, and job performance, and negatively related to employees' intention to quit and transactional contract. As expected, transactional contract is positively related to intention to quit and negatively correlates to job satisfaction and job performance. Stigma consciousness is positively related to intention to quit and negatively related to job satisfaction and job performance. Moreover, job satisfaction is negatively related to intention to quit and positively related to job performance. In summary, the correlations revealed in these results confirm our theoretical expectations.

Results
In order to determine the appropriate level of analysis for the data in Study 2, this study follows the suggestion of Schriesheim et al. (1995) by performing within-and between-analysis (WABA: Dansereau et al., 1984). WABA assesses variation and covariation simultaneously in variables within and between levels of analysis (Dansereau et al., 1984). The E and F tests conducted under WABA I indicate whether the variance is either between-group or within-group (dyad), or whether both (equivocal) or neither within-or between-group levels are the appropriate inference (null). In WABA II, A and Z tests are used to determine at which level the covariance occurs. Finally, the inferences from WABA I and WABA II are both assessed, and the within-and between-group correlation components are examined by an A test (Yammarino & Markham, 1992).
Twenty-eight supervisors participated in this study, and each supervisor rated an average of four subordinates per group. Following Dansereau et al. (1984), this study conducted WABA to test the assumption of independence among the ratings that each supervisor provided for multiple subordinates. The E-tests are all significant (the E ratio for task performance is .52 and contextual performance is 0.37, 30° test), suggesting that the variation within groups is significantly greater than the variation between groups. Further findings indicate that the average of within-groups eta (the average η W for task performance = 0.65 and the average η W for contextual performance = 0.73) is larger than between-group eta (η B for task performance = 0.34 and the average η B for contextual performance = 0.27). Using a corrected F-test to test the significance of the effect (Dansereau et al., 1984), the results show that none of the corrected F-tests are statistically significant (the average corrected F(27, 104) = .51 for task performance and F(27, 104) = .39 for contextual performance), suggesting that there is not significantly more variation within rather than between groups. The WABA results thus support the assumption of the independence of the supervisors' ratings of their subordinates' job performance. Consequently, this study conducts the analysis at the individual level by using a hierarchical regression analysis.
Moreover, based on suggestions in previous studies (e.g., Lee & Allen, 2002;Raja et al., 2004) this work uses five control variables: gender, tenure, job status, type of job, and position, which can all be related to work status (e.g., McDermott et al., 2013;Wilkin, 2013). This study also controlled the type of organization. Specifically, this study controlled for tenure (in years), demographic variables (gender was coded '0ʹ if the respondent was male and '1ʹ female), job status (permanent worker was coded '1ʹ, and temporary worker was coded as '0ʹ), type of job (administrative was coded '1ʹ and production was coded '0ʹ), and position (worker was coded '1ʹ, supervisor was coded '2ʹ, and manager were coded '3ʹ). Table 4 presents the regression results of the expected relationships.
In order to test the moderating effects, this study uses hierarchical regression, since the research variables are measured using continuous variables. The use of regression can retain the continuous nature of the variables without losing information or reducing the power to detect the interaction effects (e.g., Aiken & West, 1991;Cohen et al., 2003). However, there is the possibility that variables might correlate with each other (high multi-collinearity), and thus this study applies the centering method to reduce these effects (Frazier et al., 2004). To provide an illustration of a significant moderating effect, the procedures of Aiken and West (1991) and Cohen et al. (2003) were used, and these are in Figures 1 and 2. Hypothesis 3 predicts that stigma consciousness moderates the relationship between relational contract and job related outcomes, such that individuals with predominantly hold relational contract will have weaker positive relationships with job satisfaction, task performance, and contextual performance when they have higher stigma consciousness, but a weaker negative relationship between relational contract and intention to quit. The proposition that the negative relationship between relational contract and intention to quit tends to be weaker when respondents have high stigma consciousness is not supported in Study 1, while in Study 2 the results indicate that the moderating effect is weaker (β ¼ À 0:246; p<0:05 À M4) compared to the direct relationship (β ¼ À 0:501; p<0:001 À M4). Note: þ represents p < .10, * represents p < 0.05; ** represents p < 0.01, *** represents p < .001 The fourth figure of Study 2 indicates that respondents tend to have the lowest intention to quit when they predominantly hold relational contract and are less stigma consciousness ( � X ¼ 2:150), compared to those with more stigma consciousness ( � X ¼ 2:665). However, individuals with less relational contract and more stigma consciousness have the highest intention to quit ( � X ¼ 4:450), much greater than those who have less stigma consciousness ( � X ¼ 2:831), and thus H 3a is confirmed in Study 2.
The results also indicate that the positive effect of relational contract is negatively moderated by stigma consciousness with regard to job satisfaction (Study 1: β ¼ À 0:171; p<0:001 À M7), while the results in Study 2 show that the relationship between relational contract and job satisfaction is weaker (β ¼ 0:333; p<0:001 À M9) than the direct one (β ¼ 0:437; p<0:001 À M9). The first figures of Study 1 and Study 2 indicate that job satisfaction reaches the highest level when individuals have relational contract and at the same time have low stigma consciousness (Study 1: � X ¼ 6:912; Study 2: � X ¼ 6:171). When employees have high relational contract and at the same time they are highly stigma conscious, their level of job satisfaction dramatically decreases (Study 1: � X ¼ 5:587; Study 2: � X ¼ 5:964). In a similar manner, the level of job satisfaction for individuals who have low relational contract and less stigma consciousness (Study 1: � X ¼ 5:361; Study 2: � X ¼ 5:750) is higher compared to individuals who are more stigma conscious (Study 1: � X ¼ 4:999; Study 2: � X ¼ 438). Therefore, H 3b is supported in both studies.
There is no moderating effect of stigma consciousness on task performance, but the relationship between relational contract and contextual performance is weaker when it is moderated by stigma consciousness (β ¼ 0:298; p<0:05 À M12) compare to the direct relationship (β ¼ 0:548; p<0:001 À M12). The figure 3 shows that individuals who have high relational contract tend to have higher contextual performance when they are less ( � X ¼ 6:710) rather than more stigma conscious ( � X ¼ 5:776). In the case of individuals who are categorized as having low relational contract, different levels of stigma consciousness have small effects. Specifically, individuals who are less stigma conscious have higher contextual performance ( � X ¼ 5:474) compared to individuals that more stigma conscious ( � X ¼ 4:712). Therefore, H 3c is not supported while H 3d is confirmed in Study 2.
Hypothesis 4 posits that perceived stigma consciousness moderates the relationship between transactional contract and job related outcomes in such a way that a positive relationship between transactional contract and (a) intention to quit is strongly related, but the negative relationships between transactional contract and (b) job satisfaction, (c) task performance, and (d) contextual performance are also stronger. The regression results show that the expected moderating effect of stigma consciousness exists between transactional contract and job satisfaction in Study 1, in which the moderating effect is stronger (β ¼ À 0:211; p<0:001 À M6) than the direct relationship between the two variables (β ¼ À 0:149; p<0:05 À M6). The second figure of Study 1 indicates that there is a slight difference between high versus low transactional contract when employees are less stigma conscious ( � X ¼ 6:190 vs. � X ¼ 6:083). However, individuals tend to have low job satisfaction when they are categorized as having low transactional contract and at the same time they are more stigma conscious ( � X ¼ 5:828). The level of job satisfaction is even lower when highly transactional contract oriented individuals have higher expectations of being discriminated against by others ( � X ¼ 4:758), and thus, only H 4b is supported.
As discussed before, job status could directly or indirectly influence job related outcomes (e.g., De Cuyper et al., 2008;Stamper & Van Dyne, 2001), thus additional analyses need to be carried out. Even though temporary workers ( � X ¼ 4:83) have higher levels of stigma consciousness than permanent ones ( � X ¼ 4:13) to a significant degree (F = 12.24, p < .001), the results indicate that job status has no direct effects on the dependent variables, which is similar to the findings for other control variables. The indirect (moderating) effect was obtained based on data from Study 2. ∆R 2 of the moderating effect of job status on the relationship between relational contract and satisfaction is 0.01 (∆F = 2.57, p = 0.11), ∆R 2 for intention to quit is 0.00 (∆F =0.63, p = 0.43), ∆R 2 for task performance is 0.00 (∆F = 0.08, p = 0.77), and ∆R 2 for contextual performance is 0.01 (∆F = 0.64, p = 0.43). Similarly, ∆R 2 of the moderating effect of job status on the relationship between transactional contract and job satisfaction is 0.00 (∆F = 0.99, p = 0.32), ∆R 2 for intention to quit is 0.00 (∆F =0.02, p = 0.90), ∆R 2 for task performance is 0.01 (∆F = 0.47, p = 0.49), and ∆R 2 for contextual performance is 0.00 (∆F = 0.08, p = 0.78).

Contextual Performance
Relational Contract

Discussion
The findings of Studies 1 and 2 validate the hypothesis that relational contract has a positive relationship with job satisfaction, task performance, and contextual performance. The positive relationship is even greater for contextual performance, as relational contract oriented individuals include social-socioemotional factors in their exchanged resources, which will maintain or improve the social and psychological context within which core tasks are performed (Borman and Motowidlo, 1993). As predicted, a negative relationship was found between relational contract and intention to quit. These findings are consistent with those of Raja et al. (2004) and Millward and Brewerton (1999), who found that relational contract has a positive effect on job satisfaction and a negative effect on intention to quit. Moreover, this study extends prior research findings which showed that relational contract could positively influence extra-role-contextual (Hui et al., 2004) and in-role-task performance. However, even though the expected negative signs on the relationship between transactional contract and job related outcomes (and positive signs on intention to quit) were found, these are not significant.
This study argues that the dominant relationships between relational contract and job related outcomes could be related to the cultural context within which individuals are raised. According to Hofstede (1980), Indonesian employees can be categorized as collectivists, while a GLOBE study also indicates that the group and institutional collectivism scores of Indonesians is above the world average (House et al., 2004). Thomas et al. (2003) noted that cultural differences influence employees' perceptions concerning the terms of the exchange relationships they have with their employers. Both studies (Hofstede, 1980;House et al., 2004) indicated that Indonesian employees highly value and practice loyalty, cohesiveness, and collective action, and thus it is not surprising that they might predominantly hold relational contracts rather than transactional ones. Further, Thomas et al. (2003) also argued that employees in collectivist countries tend to have a greater tendency to predominantly hold relational contract, because it is consistent with their mental structures that cognitively process each socio-emotional cue they receive, and at the same time they are more motivated to create long-term relationships (Tse et al., 1988;Yang, 1993). On the other hand, employees in individualist countries are likely to interpret information in economic terms, which have short-term relationship orientations (Erez & Earley, 1993;Triandis, 1989), and thus such employees tend to have a more transactional contract orientation.
Furthermore, Nisbett et al. (2001) argued that East Asians tend to be cognitively holistic by considering the whole of a phenomena and assigning causality to it. They also posited that this kind of holistic cognition processes information in order to maintain harmony within it. This cognitive processing style is consistent with the views of relational contract oriented individuals, who see their exchange with the organization from many perspectives in order to develop longterm relationships based on harmony. Moreover, Chiu (1972) proposed that holistic oriented individuals are socio-oriented or situation-centered, and are thus obliged to be more sensitive to their environment. Since relational contract oriented individuals perceive that their exchanged resources are more social-socioemotional and economic in nature (Foa & Foa, 1974, 1980, the use of holistic cognitive processes is thus justified. On the other hand, Westerners who are culturally rooted in Greek traditions tend to use analytic cognition that focuses on the primary object, as well as the formal rules embedded within this way of thinking. Based on this, it can be expected that the respondents in this study would be highly likely to be relationally rather than transactionally contract oriented than would be found in a Western context.
The second finding indicates that stigma consciousness negatively moderates the effect of psychological contract. Individuals who predominantly hold relational contract tend to reduce their job satisfaction when they are highly stigma conscious. Further findings also indicate that employees' contextual performance dramatically decreases when they are high in stigma consciousness and the effects are more pronounced for lower levels of relational contract. The opposite pattern applies for individuals who predominantly hold transactional contract, as their job satisfaction reduces dramatically when they are highly stigma conscious. As discussed previously, individuals who have high stigma consciousness tend to disengage-either psychologically or physically-from the organization that they expect will discriminate against them . Consequently, the relationships between relational contract and job-related outcomes are weakened while the relationships between transactional contract and these outcomes are strengthened.

Managerial implications
Many studies have found that individuals from collectivist cultures view the self as interdependent with selected others (Triandis, 1995), and in-group membership is a central aspect of identity (Hofstede, 1980). Consequently, being different or perceived by others as different heightens the perceived threats of shame and embarrassment that these individuals feel (Kobayashi et al., 2001). Since individuals who are high in stigma consciousness expect that they will be discriminated against by others (Pinel, 1999), this condition could reduce their self-worth. According to Kim (1994), in-group exchanges in collectivist cultures are based on equality. In order to be equal with regard to the exchanged resources, highly stigma conscious individuals tend to reduce their positive attitudes toward the job and extra-role performance.
Furthermore, even though there are significantly different levels of stigma consciousness between permanent and temporary employees, the findings of this work indicate that this has no moderating effect on the relationship between psychological contract and job related outcomes. Further results also indicate that job status has no significant relationship with job related outcomes. These findings are in line with those of Conway and Briner (2002) and Coyle-Shapiro and Kessler (2002), which reported that the relationship between psychological contract and job related outcomes is not moderated by job status. The findings also support the proposition of Pinel (1999) that stigma consciousness is not based on the status or other attributes that engender discrimination from major groups per se, but more on individuals' expectations that they will experience discrimination. Moreover, the findings are also in line with the notion of Crocker et al. (1998) that carrying a "minority" status does not mean that individuals will be stigmatized by others.
In addition, previous studies mainly examined these research issues by collecting data from private organizations (e.g., Hui et al., 2004;Millward & Brewerton, 1999;Raja et al., 2004) and few used public organization (e.g., Coyle-Shapiro & Kessler, 2002McDermott et al., 2013) to explore the topic of psychological contract. This study employed two different organizations and found out that respondents from the private organization have a higher intention to quit than respondents from the public organization. This is understandable, since employees in the private organization have greater opportunities to switch to other firms that offer more benefits than the current one (Turnley & Feldman, 2000). In addition, according to Alatrista and Arrowsmith (2004), most employees of public organizations' join them with a strong sense of altruistic mission, so the resulting employment relationships have been shown to differ from those in for-profit organizations. This is also consistent with the finding in Cunningham (2010) that employees in public organizations have greater resilience in the face of difficult work conditions, as they often face funding cuts from the government.

Theoretical contributions
The growing literature on psychological contracts has paid a great deal of attention to the antecedents and consequences in relation to favorable (i.e., higher job satisfaction and lower intention to quit-e.g., Raja et al., 2004) and behavioral outcomes (OCB-e.g., Hui et al., 2004). This study empirically contributes to the psychological contract literature by confirming that relational contract positively influences job performance (both task and contextual performance), while the opposite is true for transactional contract. Moreover, this study extended its investigation by proposing a moderating variable, stigma consciousness, based on the insights of social exchange theory (Blau, 1964;Gouldner, 1960;Rai & Agarwal, 2018). The findings indicate that stigma consciousness has significant moderating effects on the influences of psychological contract on job satisfaction and job performance. Therefore, as suggested by Boyce et al. (2007), by treating all individuals in a similar manner and working against any potential stigma, such as that directed at temporary workers, managers can reduce the negative effects associated with this.
Further, this study examines two different organizations in Indonesia in its discussion of about psychological contract and stigma consciousness, and thus offers a different perspective to those works carried out in Western (e.g., De Jong et al., 2009;Scheel et al., 2013) or Chinese contexts (e.g., Hui et al., 2004;Z-X. Chen et al., 2008).

Limitations and future research directions
Although the results of the two studies carried out in this work are compelling, several limitations deserve comment, as they suggest directions for further research. The cross-sectional research design, the composition of the sample, and the response rates all serve to temper the results of this work, and several extensions would be beneficial. First, although this research uses two different organizations in two different industries, enlarging the sample to include multiple organizations and multiple-industries could increase the generalizability of the findings. Second, this study focuses on Indonesian organizations, which might imply that the results have only limited scope. This study contributes to the psychological contract literature by utilizing an Asian context (e.g., Agarwal & Gupta, 2018;Kutaula et al., 2020), while most previous studies were conducted in a Western one. However, because Asian societies place more emphasis on harmony in relationships, Asian people are expected to value relational forms of employment more highly (Cooke et al., 2020;Kutaula et al., 2020). This implies that they tend to predominantly hold relational rather than transactional contracts, and this might be very different to what is found with Western people. Thus, contrasting the effects of psychological contracts between Western and Asian contexts could be a fruitful avenue for future research. Third, the use of the short versions of the psychological contract and stigma consciousness questionnaires based on the pre-test results was not a good decision. Although it was based on the suggestions of the Vice President of HR and leader of a labor union in Study 1 and face validity was not a concern, future studies should use the original versions of these instruments to increase the validity of the results. Finally, we mainly focused on the two common psychological contracts, relational and transactional; to examine the effects on job related outcomes. We disregarded transitional contract because the context of our study at that time was stable, while Hui et al. (2004) used Rousseau's (2000) Psychological Contract Inventory and determined that transitional contract reflect a breakdown or absence of an agreement between employees and employers due to unstable environments. In the case of balanced contract, this involves "mutually satisfying relationships with open-ended arrangements with the transactional feature of well-specified performancereward contingencies" (Hui et al., 2004, p. 312), while De Jong et al. (2009 states that it constitutes as "many promises or high fulfillment of promises on the part of both employees and employers" (p. 331). When the current study was undertaken, performance-based systems were not as well established as now, and thus " . . . well-specified performance-reward contingencies" (Hui et al., 2004) did not exist. Each employee receives a similar amount of money based on conventional HR policies, such as work experience and education. However, future studies may extend our findings by employing transitional and balanced contracts (Hui et al., 2004), or examine the dichotomy of promise-based and fulfillment-based exchanges (De Jong et al., 2009) and see how this relates to stigmatization in the work place.

Conclusion
This study investigates the relationship between psychological contract and job related outcomes as well as for whom the relationship between two is stronger/weaker. Based on surveys in two different organizations, the results indicate that respondents who predominantly hold relational contract tend to have less intention to quit and are more satisfied with their current job, while also having higher task and contextual performances. Even though respondents who predominantly hold transactional contract saw the opposite effects to those who hold relational contract, the differences are not significant, and this may be due to the cultural context in which the study was carried out. The results also indicate that the perceived stigma consciousness of the respondents regarding about their job status tends to weaken the positive relationship between relational contract and job satisfaction, as well as contextual performance, and strengthen the negative relationship of relational contract with intention to quit. In the case of transactional contract, the negative relationship with job satisfaction is stronger when employees have high stigma consciousness. These findings indicate that managing