Prediction of guilt and shame proneness based on disruption to psychological contract: A new light for corruption prevention.

Amid controversy over plurality and contestation of the meanings of corruption, previous reviews and studies showed that proneness to moral emotions, i.e. shame and guilt, can predict one's corruption behavior. To give a theoretical basis for the efforts of preventing corruption that is thick with emotional nuance, this present study employs disruption to psychological contract, i.e. psychological contract breach (PCB), as a predictor of moral emotions proneness. The study involving 265 employees (169 males, 96 females; M age = 32.32 years old; SD age = 7.28 years) of four big private banks in Jakarta, the capital city of Indonesia, shows that PCB-with noting that, in this study, its scale operational scoring represents, reversely, the contract fulfillment-can predict Guilt-negative behavior evaluation (Guilt-NBE), Guilt-repair (Guilt-REP), and Shame-negative self-evaluation (Shame-NSE); all in negative directions, proved via simple linear regression analyses. Further analysis showed a more dynamic relationship between PCB and Guilt-NBE that fits to a cubic regression model. This study contributes to the axiological aspect of business psychology, especially in the ethical psychology of banking industry.

Introduction

Corruption takes place in the banking industry, a place where society should feel secure and can entrust their assets. Banking businesses with a strong element of trust needs to guarantee their clean reputation in front of their stakeholders, i.e. that their members are free from corruption.

Meanwhile, Indonesia is referred as a nation whose political elite is shameless (e.g. Anugrah, 2018) which is shown by the various behaviors of public officials arrested by Indonesia's Corruption Eradication Commission, such as smiling, laughing, even conveying signals of pride and low fear before the law through their body language (metal regards, etc.) in front of the television camera (Hakim, 2016; Suhardi, 2018). This situation could not be separated from the historical context of Indonesia, in which for more than 30 years, this country was ravaged by a “culture” of corruption. In this regard, Anwar (2016, p. 149) stated, “In the past three decades, the country had witnessed a continuous and systematic rape of its wealth and resources and a shameless willingness to throw all principles to the winds in the name of self-interests by those very people and institutions that were supposed to protect them.”

Faisal Basri (as cited in Hamdani, 2019) stated that the condition of private banks that were “imaged as being seriously financially ill” could be an instrument of corruption modus operandi for State-owned (in Indonesian: BUMN/Badan Usaha Milik Negara) banks through counterfeit rescue steps. Private banks in Indonesia have also been being used by public officials in Indonesia to save money from money laundering and corruption (Norhadini, 2020). In those cases, the banking sector that is closely related to the cycle of money seems to open big opportunities for corrupt people. Other corruption cases that may happen in banking sector include (1) Signature forgery; (2) Data forgery for granting credit (fictitious credit as a result of bank employees’ collusion with a debtor); (3) Fake letter making; (4) Violation of funding procedures; and (5) Misappropriation of customer funds involving branch manager, intermediary party, and fund placement authorities (Litbang KPK, 2007).

Those cases are only some examples of corruption cases revealed by the public and reported in mass media. One of the resource persons of the preliminary study revealed that there were some other corruption cases taking place in banks, such as price fraud in the procurement of operational equipment and project in a company. However, this kind of case is often resolved internally (e.g. with imposing a work termination process) and is not published in mass media. However, the exposure of banking corruption on society can lower society's trust in the bank, risking the bank's reputation. The act of corruption in the banking industry is also often unidentified even by the internal party and is covered-up.

More intensive and repetitive behaviors of corruption, and greater tolerance of corrupt behavior, are found in people with a low level of guilt and shame. This has been proven by the results of studies in neuropsychology, law, religiosity, and culture fields (de Sardan, 1999; Fan et al., 2020; Pakpahan, 2016; Supeno, 2020).

The guilt referred to in this present study is moral guilt, not dispositional guilt. Operationally, the differences between the two are,

“Moral guilt is .... the anxiety triggered by an enraged conscience, following an action. This anxiety is the psychological result of individuals turning their own aggression back against themselves, usually directed by their conscience .... It is tied directly with a breach of morality …. A person acts, or contemplates action. The act contravenes their conscience—their sense of what they ought and ought not to do …. [Meanwhile,] dispositional guilt, by contrast, is anxiety that is deeply embedded in character .... It precedes the development of the moral faculty—conscience ... It is what Freud usually termed unconscious guilt” (Carroll, 2020, p. 8–9).

The difference with shame—although both shame and moral guilt are self-conscious emotions—is that shame is an emotion that comes from an assessment of one's overall bad self (not just bad action or behavior) because of transgression towards moral standards, triggering defensive reactions, self-isolation from the public (not private matters), or exemplification actions, damages restitution, and apologies, as the remedy efforts to maintain positive self-image (Dasborough et al., 2020; Greenbaum et al., 2019).

Moral emotions can be generated by situations of agreement, formally or informally, between an employeeand the organization in which he/she works. For example, Morrison (1994, p. 359) stated,

“Psychological contracts add predictability. Predictability is probably the most important issue for human relationships .... Because of this human need for structure, people want to see their leaders as reliable .… If people begin to believe their expectations were unrealistic (so they were foolish to have them), shame results.”

In other words, if an employee has realistic expectations on his/her organization (note that everyone thinks that he/she is a rational being; Aronson et al., 2014), and the organization disrupts his/her hope or perceived psychological contract, then the shame of the employee will diminish, and the person will be vulnerable to do corruption.

The contract problem is historically indeed very close to corruption problem (Aji, 2018; Mahfud, as cited in Indonesia Lawyers Club, 2018, minute 12:30–15:35; Triyono et al., 2016). In a sociological perspective, morality is indeed a social agreement based on the self-interest of the parties who believes and promises that one will not do things that harm each other so that the self is not hampered/disadvantaged by others (social exchange); provided that the agreement is perceived fair (Boucher and Kelly, 1994; Juneman et al., 2012). In other words, from the time dimension, in a psychological contract, “future exchange or reciprocity” and “relational or contractual obligation” are the objects of both implicit and explicit agreements (Kingshott et al., 2020, p. 3).

The question that arises is, how is the relationship between psychological contract disruption and moral emotions? There are two explanations. The first explanation, organizational injustice resulted from perceived disruption of psychological contract produces unpleasant moral emotions (Ford and Huang, 2014)—i.e. “Emotions that concern the interests of society whether or not one's own interests are implicated” (Haidt, as cited in Ford and Huang, 2014, p. 39). Through this mechanism, employees who experience contract breaches will experience angry and will carry out justice restoration or correction or retaliation. The emphasis of the study of Ford and Huang (2014) was moral emotions that are other-directed (angry, contempt, disgust), while this present study emphasizes self-condemning moral emotions (shame, guilt) that have an impact on others or the environment, such as the organization. i.e. permissiveness towards corrupt behavior (Abraham and Pane, 2014; Smith-Crowe and Warren, 2014).

The second explanation comes from Kingshott et al. (2020, p. 3), that “Since the social exchange relationship is laden with emotions, such breaches [or disruptions] are often followed by perceived violations that comprise a negative emotional stated directed towards organization.” Interestingly, Kingshott et al. claimed through their research article that this cognition of social exchange applies across cultures, both in Western and Eastern societies. Even in Eastern societies who are collectivistic and have high conflict avoidance, such as Indonesia (Panggabean, 2004), understanding of social exchange in psychological contracts is becoming increasingly urgent. This is because in such culture, (1) expectation in an exchange, which is a function of cultural norms, is not always clear, and thus, (2) the consequences of perceived breaches or disruptions become more eroding, more damaging in the form of actions that thick with emotional shades, namely “destructive passive action of neglect” (Kingshott et al., 2020, p. 9).

Based on those explanations, the corrupt or disrupted contract implies, therefore, bad morality. This present study assumes that it is very pivotal to empirically investigate at the socio-psychological level, contract's predictive power over one's morality.

For practical purposes, in this article, the term “disruption” has the connotation of negative disruption and is synonymous with breach and violation. Accordingly, the three terms are used interchangeably. The conditions for disruption are the attribution of responsibility (that the contract transgression is carried out by external parties outside of their selves, in this case, the company, especially by intention), contextual explanation (that there is a comparison between employee contributions—actual and potential—and their company's reciprocation), and attempt to clarify (that requests for explication about their perceptions regarding violations have been ignored, or not reasonably responded to, by the company) (Parzefall and Coyle-Shapiro, 2011).

Literature review

In the socio-psychological level, there is a trend in social exchange perspective (reciprocity norms) explaining employee's expectation and attempt (not necessarily shared with his/her employer) to balance mutuality and justice in (1) employee's right, (2) employee's contribution, and (3) employer's obligation. The three things mentioned are in the “agreed” perceived promises, usually informal (implicitly or explicitly based on employer's act) in a special situation (not related to personality traits), go beyond written or legal contract called “psychological contract” (Johnson & O'Leary-Kelly, 2003). Disruption to psychological contract, or psychological contract violation takes place when there are (1) incompatibility or distance of the expectation between employers and employees, or (2) experience of betrayal or perceived broken outcome based on shared expectation between them, both intentional (deliberative, or employer's willful violation) and unintentional (as a result of disruption or unexpected environmental change afflicting employers).

So far, many effects of psychological contract breach or disruption (PCB) in the non-moral area (i.e. no explicitly moral stressing in the research findings), such as absenteeism, job satisfaction, citizenship behavior, turnover, commitment, work attitude, and organizational-level performance, are investigated. The investigations were done both directly or interactively with other variables, e.g. quality of an existing person-organization relationship, working social exchanges, post-violation breaking honest explanation, intentionality attribution interacted with position level of employee, measurement method (composite vs global), and contract content (relational vs. transactional) (Bal et al., 2010; Suazo et al., 2005; Zhao et al., 2007), which are mediated by personal distrust of perceived employer's integrity (which threatens one's identity values) (Johnson & O'Leary-Kelly, 2003). The negative effects sourced from the acts of personal restoration to create a balance and restore distorted justice perception, i.e. by doing defective actions or decreasing cooperative attitudes toward the organization (Bal et al., 2010; Suazo et al., 2005).

Previous studies indicated that anger (and behavior tendency associated with it) is an emotional reaction accompanying psychological contract violation especially if the cause of the violation is attributed to controllable matters (Morrison and Robinson, 1997). Anger is a part of moral emotions, especially social-demeaning emotions (besides disgust, contempt) (Russell and Giner-Sorolla, 2013).

To the best of our knowledge, there has not been any researcher who investigates the predictive relationship between psychological contract breach and self-conscious moral emotions, which are guilt and shame, even though the proneness of the two moral emotions can predict corrupt behavior (Abraham and Pea, 2020; Cohen et al., 2011), one of the problems infecting Indonesia and many other developing countries. However, there was a study that is closely related to this present study. Barclay, Skarlicki, and Pugh (2005) did not conduct a special investigation on psychological contract and did not even mention the term in their research article, but they found that “When either procedural or interactional justice was low, employees reported relatively low levels of shame and guilt notwithstanding the favorability of their outcome” (p. 637). As stated in the Introduction, perceived injustice is one of the psychological phases through which a person experiencing contract disruption/breach/violation.

An individual's tendency to do corruption is closely related to the moral emotion of the individual. Individuals with higher moral emotion proneness will be more careful in decision making that violates moral principles. Moral emotion is an important key in shaping an individual's moral behavior and is a bridge influencing how individuals meet moral and behavior standards (Tangney et al., 2007). Therefore, the clarification on the relationship between psychological contract breach and inward-focused moral emotions gives a practical benefit to prevent corrupt behavior. This present study contributes to the matter.

There were some reasons why moral emotion proneness was emphasized in this present study. Lately, it has been found that emotional intelligence can be the buffer against (as a moderating variable that desensitizes) the negative effect of PCB on workplace deviance (Balogun et al., 2018). Therefore, predicting moral emotion deserves to be a priority because it is the human's emotional dimension that can determine moral behavior in a concrete situation. Balogun et al. (2018) showed that emotional intelligence—as a personal disposition that can recognize, understand, and manage emotions and stimuli related to emotion—may prevent negative emotion heuristic produced by PCB. The prevention was done through “disconnecting from anger rumination and hostile revenge thoughts or replacing them with positive mood” mechanisms both in the context of “economic depression” and “workplace relations” (Balogun et al., 2018, p. 9, 12). In this case, emotional intelligence has a compensatory effect. However, this present study assumed that moral emotion proneness had facilitation effect—not compensatory—on negativity caused by PCB and brought it to the morality domain. PCB decreased one's proneness to put forward the role of moral emotion to anticipate, decelerate, and regret immoral behavior and self experiencing moral erosion.

Theoretically, it is reasonable to relate psychological contract breach and susceptibility to moral emotions. Montes and Zweig (2009) stated the possibility of applying social comparison theory in the analysis of dynamics of psychological contract breach. However, people contrast gain and expectations—concerning their wants—and this contrasting effort are formed and shaped by experience or social evaluation. Meanwhile, Noerhardiyanty and Abraham (2015) reported that the comparison decreases the likeliness of being prone to shame as noble ethical sentiments, consisting of negative self-evaluation and withdrawal tendency. With the combination of two propositions, this study hypothesized that psychological contract breach has a negative correlation with proneness to moral emotions, i.e. guilt and shame.

Operationalizing the hypothesis: When one who feels that his/her psychological contract is violated, it will be easier for the individual to misappropriate. Disappointment resulting in the emergence of rationalization process—that the misappropriation is not an immoral action but an effort to “balance” the situation—encourages these individuals to (1) be more reluctant to negatively judge misappropriate behavior they do, (2) to lower their striving to improve their behavior when they do misappropriation, (3) not feel guilty and have lower negative self-evaluation, and (4) have a bigger possibility to withdraw themselves from any responsibility for their mistakes.

To explain, moral emotion has four dimensions, namely negative behavior evaluation (Guilt-NBE), initiative to fix errors (Guilt-REP), negative self-evaluation (Shame-NSE), and withdrawal behavior (Shame-WIT) (Cohen et al., 2011). Guilt-NBE measures ones' moral and emotional disposition to regret a violation, while Guilt-REP evaluates his/her behavioral orientation to correct and apologize for wrong actions (Cohen et al., 2011). Furthermore, Shame-NSE depicts emotional responses to regrets on publicly exposed transgression, while Shame-WIT illustrates an action orientation to hide, withdraw, or avoid the consequences of an ethical violation. Tangney (as cited in Cohen et al., 2011) explained that moral emotion gives strong encouragement for individuals to do good things and avoid bad attitudes. The measurement of moral emotion is an effective method to detect employees’ tendencies to be corrupt.

Shame is often perceived as an emotion similar to guilt. Nevertheless, these two moral emotions have significant differences. Guilt is regret occurs when making mistakes even though the mistake is not noticed by others. The regret emerging from guilt is a personal thought of a specific adopted morally bad behavior, not depicting judgment on the individual's self as a whole. Smith et al. (as cited in Cohen et al., 2011) described guilt as an emotion emerging when individual acts contrary to his/her conscience. Different from that, shame emerges when the mistake made is revealed to other parties, or when individuals worry that their mistakes will be exposed to the public (Cohen et al., 2011).

When employees are frustrated because they perceive a contract breach by the company, employees will respond by doing behaviors disadvantaging the company (Bordia et al., 2008). This is an effort to balance the condition and injustice created by the company. The frustration encourages moral emotion shift that includes shame and guilt tendencies in employees’ behavior (moral disengagement, moral looseness). The frustration will inhibit employees (1) to empathize with the loss experienced by the company and (2) to fix errors made because the flaws are perceived to come from the company. The negative emotion will provoke employees to run away from the responsibility for the consequences triggered by their behaviors.

Those explanations could be complemented with Alcover et al. (2017, p. 4, 9) study finding that psychological contract “reduce(s) uncertainty between the parties (e.g., by defining roles and specifying future courses of action) .... [is] a sense-making process associated with organizational socialization.” This present study widens the applicability of Alcover et al.’s statement on morality. Moral behavior uncertainty can be anticipated by socializing information on the things that can be expected and the things that are supposed to be fulfilled so that mutualism produces positive morality in the workplace.

Further, O'Donohue and Nelson (2009) once indicated that employees' reciprocal contribution towards psychological contract fulfillment from the company can be in a form of “soft” or non-material like ideological contribution, such as affinity and advocacy of abstract moral values or principles considered as life guideline. This present study assumes that this soft contribution engulfs (becomes the spirit of) exchanges in the psychological contract. Psychological contract orientation is not only transactional (focuses on utility contract for employees, that is individualistic), or relational (focuses on relationships among the parties perceived to be “under the term of” contract contents, that is collectivistic), but also transpersonal, that is “what fits with me, how do we work together in the organization, and where is the fit with me, us, and the rest of society” (Burr & Thomson, as cited in O'Donohue and Nelson, 2009, p. 252).

O’Donohue and Nelson (2009) study assertively showed that the common good is a layer that is inseparable from a psychological contract. Common good considers not only the sustainability of a company but also relationship sustainability between a company or an organization and the public. The level of happiness, personal development, family wealth, etc., obtained from a company as its psychological promise is certainly a built environment (including in its activities, lifestyles, local economy, social capital, and people), which is related to the biosphere and global ecosystem; see Barton (2005) ecological model. Kals and Maes (2002) indeed indicated emotion significance in morality and sustainable behavior, “It makes sense that especially social responsible subjects, or subjects with highly developed moral norms, should be willing to accept personal burdens and costs benefiting the whole society” (Kals and Maes, 2002, p. 101). Applying their statement in this present study, people with highly developed moral norms are those with high moral emotion proneness and low perceived psychological contract breach. The psychological contract effect is like a “whirlpool”. First, it will influence moral emotion (Guilt-NBE, Guilt-REP, Shame-NSE, and Shame-WIT), and after that, moral emotion proneness will lead to a transpersonal sense of psychological contract that makes one willing to lose—provided that the society is benefited.

Psychological contract represents the interaction between self and social situation

Studies of moral emotions suggested that the emotions were influenced by the self (e.g. emotional intelligence, Balogun et al., 2018; “self-importance of moral values”, Johnston and Krettenauer, 2011) or society (e.g. social control, Harris, 2003). This present study emphasizes that moral emotions are influenced by perceived psychological contract that is the locus of interaction between the self and society represented by employees and employers as well as their institutions/companies.

Previously, the interaction effect between the self (person) and the social context (context, e.g. bystander or witness existence and identities, victim's emotional reaction) on moral emotion was investigated by Roos et al. (2011) in aggression cases of children towards their peers. Using “contextualist approaches to personality” paradigm (Roos et al., 2011, p. 688), Roos et al. found that moral emotion (shame and guilt) development contributing to moral (dis-)engagement was influenced by the interaction; specifically, aggression reduction was supported by well-acknowledged friend presence and victim's sadness factors. The assumption was that children approaching teenage years favored aspiration and anticipated approval more than significant others or reference group to build positive self-identity. The finding of Roos et al. (2011) study gave important clues about the dynamics of moral emotion in facing external and internal factors.

The generalization about their study finding on the hypothesis of this present study may be limited because their participants were children, whereas the participants of this present study were adults. However, a longitudinal study of Krettenauer et al. (2013) managed to expand the actualization of the generalization potential. They found, in general, the presence of “long-term relations between moral emotion attributions in childhood and adolescence and antisocial conduct in early adulthood ... [that is] independent of the effect of conscientiousness and agreeableness and independent of the effect of aggressiveness in childhood”, even though they also found instability of moral emotion attribution especially when children experience intensive cognitive development (Krettenauer et al., 2013, p. 197, 198). There are some points to learn in Krettenauer et al. (2013) finding. The influence of interactive selfhood with situational factors on moral emotions applies to children, teenagers, and adults. Selfhood in morality is primarily formed by moral identity (Abraham and Berline, 2015; Johnston and Krettenauer, 2011). Meanwhile, situational factors contribute to the instability of moral emotion attribution or justification, and those factors are contributed by social information, such as socioeconomic-status, environments riskiness, community ethics violation as well as morality cues of significant others (Arsenio et al., 2009; Tangney et al., 2007). In this context, psychological contract being investigated covers those rich factors.

Materials and methods

This present study used a quantitative approach with survey methods. This approach did not manipulate independent variables, did not randomize participants into the experimental and control groups, and did not control the plausible extraneous variables. Therefore, although it will provide results regarding the prediction of dependent variables based on the values of independent variables, this study does not conclude the presence or absence of causal relationships between variables.

The main steps of the methods of this study were: Step 1: justification of the chosen model for analysis (covered in the Theoretical Model and Design section below); Step 2: Collecting the data (covered in the Participants and Design as well as Instrument section below); and Step 3: Analyzing the data (covered in the Data Analysis section below).

Theoretical model and design

This study adopted predictive correlational design with PCB as predictors (independent variables), and proneness to four moral emotions (i.e. Guilt-NBE, Guilt-REP, Shame-NSE, and Shame-WIT) as the criterion (dependent variables). The visualization of the theoretical model is presented in Figure 1.

Figure 1

The theoretical model. Note. Source: author own conception, based on the literature review.

This present study was initially approved by the Bina Nusantara University Research Committee, vide Letter of Consent No. 005A/Dir.RIC/IV/2013; and is a follow-up study of previous subsequent published studies of Abraham and Pradipto (2016) as well as Abraham et al. (2017)—as a series of outputs from the 2013 research project. The ethical decree is stated in Article 1 Paragraph 2 of the Letter.

Furthermore, this study, needing a larger data collection pool, is then expanded and supported by the Bina Nusantara University International Research Scheme (in Indonesian: Penelitian Internasional BINUS/PIB) with the Research Contract Letter No. 026/VR.RTT/IV/2020. The ethical decree is stated in Article 1 Paragraph 3 of the Letter. This research was also previously supported by the Directorate of Research and Community Service of the Indonesian Ministry of Research and Technology/National Research and Innovation Agency (Kemenristek/BRIN) (Higher Education Flagship Basic Research/PDUPT Research Grant Scheme) based on Letter No. B/87/E3/RA.00/2020 announced at http://simlitabmas.ristekdikti.go.id. However, because of the Covid-19 pandemic situation, the Indonesian Government through the Ministry realized the 2020 Budget Rationalization, with Letter No. B/196/M/KU.00.01/2020, by delaying research funding including for PDUPT Research Grant Scheme, for a year, which did not refocus on the Covid-19 problem.

Participants

The participants of this study were 265 bank employees (169 males, 96 females; Mage = 32.32 years old; SDage = 7.28 years) in Jakarta, the capital of Indonesia, a country where the majority of the citizens are Islam. Participants' age ranged from 23 to 54 years old, with the majority, 84.25%, were young adults (25–40 years old). Most participants possessed a bachelor's degree (224 participants) and master's degree (35 participants). The rest possessed an associate degree (4 participants) and a few others (2 participants) were not willing to give information about their highest level of education.

Participant recruitment was done in four banks in Jakarta, namely HSBC (66), BCA (67), OCBC NISP (66), and BII (66). Data on the number of employees of each bank that could be obtained from the official websites were 2,060 (HSBC; HSBC Indonesia, 2014), 22,161 (BCA; PT Bank Central Asia Tbk, 2014), 6,654 (OCBC NISP; OCBC NISP, 2015), and 5,000 (BII, 2014); or a total of 35,875 for the four banks. This number represented the number of employees of each bank nationwide. This study could not successfully obtain data on the total population of employees at the provincial level, in Jakarta, where this study was conducted. Therefore, this study implemented a non-probability, convenience sampling technique. This sampling technique is:

“the most commonly used sampling method in behavioral science research …. [in which] people are selected on the basis of their availability and willingness to respond … [It is] more timely technique than the probability sampling techniques …. [although it] offers no guarantees of a representative and unbiased sample [but not] hopelessly flawed” (Gravetter and Forzano, 2009, p. 141).

The authors contacted all their networks at the four banks via telephone, SMS, WhatsApp, and e-mail and got 265 employees (or about 7.19‰ of the total four bank employees at the national level) with maintaining a fairly balanced proportion of participants between banks.

Those banks were chosen because they were included in one or more following categories: banks with the biggest assets (BCA and OCBC NISP; Yudistira, 2018a), a bank with the most valuable brand (HSBC; Franedya, 2018), and banks with the highest net profit (BCA, Maybank Indonesia/previously BII; Yudistira, 2018b). They are regarded as the most concerned parties about the reputation of free from corruption because the business they run is a business of trust. The chosen participants were those who had worked in the banking business for at least a year. Based on the data gathered, participants had an average of 4-year experience, with the longest experience of 24 years. They were assigned by their employers to the front line, middle, and back end positions.

Instrument

To measure psychological contract breach (PCB), the author constructed a psychological scale by adapting the elements of Robinson and Morrison (1995), Turnley et al. (2003), as well as Boes (2006) instruments. The introduction of this scale were adapted into Indonesian from Boes's (2006, p. 184) questionnaire as follows:

“We are interested in how well your organization has kept the commitments it made to you. For each item listed below, please indicate how the amount that you received compares to the amount that you think vour organization should provide. Compared to what I was promised, the amount that I actually receive is:”

The items of the measurement instruments were adapted from Robinson and Morrison (1995) and developed by the author into 18 items representing the four domains of (1) job satisfaction, (2) salary and other compensation satisfaction, (3) support, and (4) career opportunities. The following are the examples of the items: “The total challenge and happiness in my job”, “The opportunity for me to develop”, “The extent for me to have regular pay raise”, “The total suggestion or opinion I receive to make a decision that gives an impact on my job”, “The quality of my company's career development program”, “The total support I receive from the company regarding the personal or family matter”, “The amount of power in my job”, “The expectation of the bonus I will receive for any additional or extraordinary job I do”, “Health benefit provided by the company”.

The response options were adapted from Turnley et al. (2003). They were: “Receive much less than promised” (scored 1), “Receive less than promised” (scored 2), “Receive about the same as promised” (scored 3), “Receive more than promised” (scored 4), and “Receive much more than promised” (scored 5). Based on the response options sequence, higher total scores on this scale indicate a person's perception that his/her psychological contract is increasingly fulfilled or not disrupted/violated.

Item validities and internal consistency reliability testing showed that, after eliminating 1 item (item number 18), the corrected item-total correlations (CIT > 0.25) ranged from 0.30 to 0.79 and Cronbach's α = 0.93 (n of items = 17).

To measure guilt and shame proneness, the authors adapted the GASP scale from Cohen et al. (2011). The instruction of this scale is as follows (Cohen et al., 2011, p. 966):

“In this questionnaire you will read about situations that people are likely to encounter in day-to-day life, followed by common reactions to those situations. As you read each scenario, try to imagine yourself in that situation. Then indicate the likelihood that you would react in the way described.”

This scale consisted of, initially, 40 items representing four dimensions, namely Guilt-NBE, Guilt-REP, Guilt-NSE, and Guilt-WIT. There were 16 items that adapted from the original items of Cohen et al. (2011), whereas the other 24 items were developed by the authors based on the four dimensions.

The examples of the items from the original Cohen's study are: “After realizing you have received too much change at a store, you decide to keep it because the salesclerk doesn't notice. What is the likelihood that you would feel uncomfortable about keeping the money?” (Guilt-NBE); “You are privately informed that you are the only one in your group that did not make the honor society because you skipped too many days of school. What is the likelihood that this would lead you to become more responsible about attending school?” (Guilt-REP); “You rip an article out of a journal in the library and take it with you. Your teacher discovers what you did and tells the librarian and your entire class. What is the likelihood that this would make you would feel like a bad person?” (Shame-NSE), “After making a big mistake on an important project at work in which people were depending on you, your boss criticizes you in front of your coworkers. What is the likelihood that you would feign sickness and leave work?” (Shame-WIT).

Original items from the GASP Scale were added with items describing the context of the office working routines, as shown in Table 1.

Table 1

Examples of items added to the GASP Scale.

Moral Emotion Proneness Dimension	Item
Guilt-NBE	You utilized a small portion of the company's operational funds due to the presence of your own essential needs. How much guilt will you feel about that act?You asked the courier in your office to deliver your personal belongings to a friend. Would you feel sorry for your actions?You caused loss of an office inventory, considered essential and expensive, but you did not admit and therefore resulting in the deduction of the salaries of all employees. How disgraced will you feel of your behavior?
Guilt-REP	At the end of the year, you pretend to be sick, in order to get permission to care for your sick parents. How considerate will you be more prepared in setting aside the amount of leave for unexpected things?When processing data, you realized you skipped one step, but no one was aware of it. How likely are you to work overtime and correct your mistakes?You postponed salary calculation for apprentice employees because of your engagement in the mall that afternoon, to buy your best friend's birthday gift. You are therefore resulting in late payment. How sensitive will you attempt to be towards the needs of the apprentice employees?There was a new co-worker who joined in your workroom. You leaked some company secrets, which spreaded to other employees as well. How likely are you to think more carefully before you speak?
Shame-NSE	Yo felt forced to feign sickness because of your inability to complete a task within a stipulated time. How likely are you to consider yourself incompetent?Within the office, you were the last person to go home, but you nudged the printer, which fell and broke. Hence, you immediately put it back to its original position as though it was not damaged. How likely are you to admit yourself as cowardice?During a lunch break, you exceeded the prescribed rest limit. Your superior observed this behavior from outside because you used a specific name tag or uniform. How bad do you feel as a person?
Shame-WIT	Your manager said that based on the data obtained, more work time you spent to build your own reputation, which were not for the benefit of the work team. How likely are you to resign from the company?Within the office, you decided not to work but engaged yourself with porn sites. Your supervisor realized this. What is the possibility that you will quit your job?You found an expensive ballpoint pen on your desk, and thus decided to use it. However, when the owner saw it, you claimed it's yours. With further investigation, it turned out the name of the true owner was imprinted on it. How likely are you to be quick to leave the situation?

Note. Source: author own conception, based on Cohen et al. (2011, p. 966).

The response options of this scale were “Very Unlikely” (scored 1), “Unlikely” (scored 2), “Slightly Unlikely” (scored 3), “About 50% Likely” (scored 4), “Slightly Likely” (scored 5), “Likely” (scored 6), and “Very Likely” (scored 7). Higher total scores on this scale indicate a person's perception that his/her proneness to moral emotions is higher. Item validities and internal consistency reliability testing showed that, after eliminating 7 items: (1) For Guilt-NBE scale, the corrected item-total correlations (CIT > 0.25) ranged from 0.27 to 0.65 and Cronbach's α = 0.82 (n of items = 12); (2) For Guilt-REP scale, the corrected item-total correlations (CIT > 0.25) ranged from 0.36 to 0.62 and Cronbach's α = 0.83 (n of items = 11); (3) For Guilt-NSE scale, the corrected item-total correlations (CIT > 0.25) ranged from 0.43 to 0.80 and Cronbach's α = 0.84 (n of items = 5); and (4) For Guilt-WIT scale, the corrected item-total correlations (CIT > 0.25) ranged from 0.31 to 0.54 and Cronbach's α = 0.71 (n of items = 7); so the total of GASP items is 35.

Data analysis

The underlying data are available at Figshare: Proneness to Moral Emotions and Disruption to Psychological Contract, accessible via https://doi.org/10.6084/m9.figshare.12110910.

Multiple linear regression analyses were done four times to investigate the predictive power of independent variables (Guilt-NBE, Guilt-REP, Guilt-NSE, and Guilt-WIT) over a dependent variable (PCB) using IBM SPSS Statistics 25 for Windows.

Results

Table 2 presents descriptive statistics and zero-order correlations between variables.

Table 2

Descriptive and correlational statistics between variables (N = 265).

	Variable	M	SD	1	2	3	4	5
1	Guilt-NBE	57.99	13.22	1	0.83∗∗	0.76∗∗	0.12∗	-0.30∗∗
2	Guilt-REP	53.60	11.08		1	0.75∗∗	0.06	-0.34∗∗
3	Guilt-NSE	23.07	7.01			1	0.15∗	-0.46∗∗
4	Guilt-WIT	24.89	7.28				1	0.00
5	PCB	45.97	11.46					1

Note. Source: author's data analysis output, based on IBM SPSS Statistics 25 for Windows. Guilt-NBE = Guilt-Negative Behavior Evaluation Proneness, Guilt-REP = Guilt-Repair Proneness, Guilt-NSE = Shame-Negative Self-Evaluation Proneness, Shame-WIT = Shame-Withdrawal Proneness, PCB = Psychological Contract Breach (i.e. the extent to which people feel that their psychological contract is not violated.). ∗p < .05 ∗∗p < .01.

The results of the linear regression assumption test showed that the data were normally distributed (see Figures 2, 3, 4, and 5) and free from heteroscedasticity (see Figures 6, 7, 8, and 9).

Figure 2

Normality test result of the prediction of Guilt-Negative Behavior Evaluation. Note. Source: author's data analysis output, based on IBM SPSS Statistics 25 for Windows.

Figure 3

Normality test result of the prediction of Guilt-Repair. Note. Source: author's data analysis output, based on IBM SPSS Statistics 25 for Windows.

Figure 4

Normality test result of the prediction of Shame-Negative Self Evaluation. Note. Source: author's data analysis output, based on IBM SPSS Statistics 25 for Windows.

Figure 5

Normality test result of the prediction of Shame-Withdrawal. Note. Source: author's data analysis output, based on IBM SPSS Statistics 25 for Windows.

Figure 6

Normality test result of the prediction of Guilt-Negative Behavior Evaluation. Note. Source: author's data analysis output, based on IBM SPSS Statistics 25 for Windows.

Figure 7

Heteroscedasticity test result of the prediction of Guilt-Repair. Note. Source: author's data analysis output, based on IBM SPSS Statistics 25 for Windows.

Figure 8

Heteroscedasticity test result of the prediction of Shame-Negative Self Evaluation. Note. Source: author's data analysis output, based on IBM SPSS Statistics 25 for Windows.

Figure 9

Heteroscedasticity test result of the prediction of Shame-Withdrawal. Note. Source: author's data analysis output, based on IBM SPSS Statistics 25 for Windows.

Simple linear regression analyses showed that:

PCB can predict Guilt-NBE (F (1, 264) = 25.95, p = 0.00, R2 = 9.00%). The details of the results are listed in Table 3.

Table 3

Simple linear regression analysis predicting Guilt-NBE (N = 265).

Variable	B	SE B	β	t	p
PCB	-0.35	0.07	-0.30	-5.09	0.00

Note. Source: author's data analysis output, based on IBM SPSS Statistics 25 for Windows. PCB = Psychological Contract Breach (i.e. in this present study measurement, the extent to which people feel that their psychological contract is not violated.).

PCB can predict Guilt-REP (F (1, 264) = 33.48, p = 0.00, R2 = 11.3%). The details of the results are listed in Table 4.

Table 4

Simple linear regression analysis predicting Guilt-REP (N = 265).

Variable	B	SE B	β	t	p
PCB	-0.33	0.06	-0.34	-5.79	0.00

Note. Source: author's data analysis output, based on IBM SPSS Statistics 25 for Windows. PCB = Psychological Contract Breach (i.e. in this present study measurement, the extent to which people feel that their psychological contract is not violated.).

PCB can predict Shame-NSE (F (1, 264) = 72.05, p = 0.00, R2 = 21.5%). The details of the results are listed in Table 5.

Table 5

Simple linear regression analysis predicting Shame-NSE (N = 265).

Variable	B	SE B	β	t	p
PCB	-0.28	0.03	-0.46	-8.49	0.00

Note. Source: author's data analysis output, based on IBM SPSS Statistics 25 for Windows. PCB = Psychological Contract Breach (i.e. in this present study measurement, the extent to which people feel that their psychological contract is not violated.).

PCB can not predict Shame-WIT (F (1, 264) = 0.00, p = 0.99, R2 = 0.00%). The details of the results are listed in Table 6.

Table 6

Simple linear regression analysis predicting Shame-WIT (N = 265).

Variable	B	SE B	β	t	p
PCB	0.00	0.04	0.00	-0.02	0.99

Note. Source: author's data analysis output, based on IBM SPSS Statistics 25 for Windows. PCB = Psychological Contract Breach (i.e. in this present study measurement, the extent to which people feel that their psychological contract is not violated.).

Discussion

It appears that when there is a stronger perception that one's psychological contract is not disrupted/violated, the lower his/her proneness to Guilt-NBE, Guilt-REP, and Shame-NSE.

However, further analysis using non-linear regression gave results that the coefficient of determination (R2)—the effect size of regression—experiences increase in the relationship between PCB and Guilt-NBE when modeled as a cubic regression (R2 moves from 9.00% to 9.80%, F (3, 264) = 10.60, p = 0.00). The increase does not occur in the relationship between PCB and Guilt- REP, Shame-NSE, and Shame-WIT.

There is a more dynamic relationship between PCB and Guilt-NBE than between PCB and Guilt-REP and Shame-NSE. Furthermore, there are three phases of the relationship (Table 7, Figure 10), which include: (1) First phase: Initially, a higher perception of the unviolated psychological contract increases the Guilt-NBE, which happens quite rapidly (steeper curve); (2) Second phase: Turning points occur at a certain level (PCB is around 30, see Figure 10) of the fulfillment of psychological contracts when a decrease in Guilt-NBE occurs slowly (more gentle curve), and (3) Third phase: After at a certain point (PCB is around 60, see Figure 10), a higher fulfillment of psychological contracts enhances the Guilt-NBE, although slowly.

Table 7

Cubic regression analysis predicting Guilt-NBE (N = 265).

Variable	B	SE B	β	t	p
PCB	4.35	2.00	3.77	2.17	0.03
PCB ∗∗ 2	-0.11	0.05	-8.63	-2.32	0.02
PCB ∗∗ 3	0.00	0.00	4.62	2.27	0.02

Note. Source: author's data analysis output, based on IBM SPSS Statistics 25 for Windows. PCB = Psychological Contract Breach (i.e. in this present study measurement, the extent to which people feel that their psychological contract is not violated.).

Figure 10

Results of cubic regression analysis in predicting Guilt-NBE (Negative Behavior Evaluation). Note. Source: author's data analysis output, based on IBM SPSS Statistics 25 for Windows. PCB = Psychological Contract Breach (i.e. in this present study measurement, the extent to which people feel that their psychological contract is not violated.).

The first phase of the relationship between PCB and Guilt-NBE could be explained using assumptions built in the Introduction. An elevated feeling of fulfillment (i.e. not disrupted) on a psychological contract causes a higher moral emotion proneness (Guilt-NBE) because justifying retributive efforts over the feeling of being mistreated is irrelevant. Therefore, the fulfillment of the PCB builds a healthy moral feeling.

The second phase of this relationship is interesting because when someone feels that the PCB is more satisfied than the expected level of fulfillment, the Guilt-NBE decreases (people no longer have a negative evaluation of unethical behavior). Furthermore, this could be explained by the overjustification effect. According to this theory, in instances where someone perceives that he/she has experienced a much greater reward, the intrinsic motivation to work for the common good decreases. Bénabou and Tirole (2006) stated:

“Rewards act like an increase in the noise-to-signal ratio, or even reverse the sign of the signal, and the resulting crowding out of the reputational (or self-image) motivation to contribute can make aggregate supply downward-sloping over a wide range, with possibly a sharp drop at zero.” (p. 1654)

Markowitz and Shariff (2012) reaffirmed that internal standards are weakened with an increase in social rewards, also known as the counterproductive effect of environmental incentives. Please note that PCB is a perception, which is also known as “the reality” for the perceiver (Diemer et al., 2015).

The third phase of the relationship between PCB and Guilt-NBE reinforces the critical review of Pittenger (1996), which stated that the overjustification effect was not invariant, with the premise that the time variable (reward schedule) and measured performance type could cause changes. However, applying Pittenger's proposition to this study, it can be assumed that research participants with perceptions about fulfilled substantial (very large) psychological contracts possess a reflection on the need to accentuate their moral standards. Hence, their accessibility to moral emotions returns to rise. Moreover, this requires further study to test why this happened. This research speculates that people who are full of psychological contracts in excessive conditions will shift their focus from money (materials) to time (more abstract investment) and this approach reduces their tendency to cheat and increases their ethicality (Gino and Mogilner, 2014; Mogilner et al., 2018).

There is a negative predictive relationship between PCB and Guilt-REP, which indicates that a substantial fulfillment (i.e. not disrupted) of a person's psychological contract leads to an enhanced weakness in correcting transgressive actions. Utilizing the sense-making theory, Chaudhry et al. (2009) explained that contextual factors could indeed change one's evaluation of PCB as either more positive or negative. Furthermore, this concept also matches the nature of Guilt-REP, which is a private sense (Cohen et al., 2011, p. 948), implying that the influence of PCB on moral emotions is also subjective to change. This context is heterogeneous and differs between individuals and institutions, therefore, even though the institutions have fulfilled the bond, the individual's perception is nevertheless dependent on entities or situations that surround the favourability. Furthermore, these circumstances also affect the social cognitive processes regarding the company, consisting of (1) justification attribution—encompassing one's perception about whether the company is considered to carry out an action that can be justified, (2) foreseeability—regarding one's consideration about whether the company is considered to take an action of which the results should have been predicted, and (3) intentionality—regarding one's perception whether the company is deemed to carry out the intended action (Chaudhry et al., 2009).

The materialism context in banking plays a role that actively raises initiatives to take advantage of deceptive legal and illegal activities (Lu and Lu, 2009). According to Lu and Lu, there was no difference in the level of unethicality between people that are Islam and non-Islam. However, office workers (like bank employees who participated in this study) were found to be more materialistic than self-employed and blue-collar workers. “A materialistic individual may be less likely to take an ethically high ground” (Lu and Lu, 2009, p. 205), while Guilt-REP is to “inhibit unethical decision making and delinquency” (Cohen et al., 2011, p. 962), which further explains the negative relationship between the fulfillment of psychological contracts and Guilt-REP.

The predictive relationship between PCB and Shame-NSE is also negative. Therefore, a more fulfilled (i.e. not disrupted) psychological contract leads to weakness in carrying out a negative self-evaluation of moral transgression. This relationship could be explained by examining the substantive difference regarding the self and acts of ethical violation, of which attribution of guilt is specific, internal, and unstable, while shame, on the other hand, is global, internal, and stable.

People with higher contented psychological contracts feel increasingly able to modify their stable and global attributes, therefore, enhancing their ability to fortify negative feelings while carrying out moral transgressions. How is this possible? Research of Gardner et al. (2015) reported a positive correlation between the fulfilment of relational (non-financial) psychological types and organizational-based self-esteem (OBSE) (r = 0.11, p < 0.05). However, in a predictive model, relational psychological contract fulfillment was also capable of predicting OBSE (β = 0.15, p < 0.05). It is worth noting that most of the PCB questionnaire items used in this research (3 of the 4 dimensions) are in the relational domain, i.e. i.e. support, job satisfaction, and career opportunities. Sivanathan, Molden, Galinsky, and Ku (2008) stated about a decade ago that rising self-esteem produces a boomerang effect, including the low Shame-NSE. Subsequently, Tremmel (2008) reviewed this statement and concluded matters that could explain this effect, which include:

“(1) Some types of positive feedback actually can escalate perceived threats to the ego and increase the need to prove that a questionable decision was the right one. (2) Across several studies, the research examines how boosting self-esteWem—whether contemplating one's own accomplishments or receiving positive feedback from others—affects the face-saving impulse to justify and recommit to decisions whose outcomes seem dubious at best.” (para 3–4)

Baumeister et al. (2003) gave an instance of Adolf Hitler, who had high self-esteem but low ethical behavior. Hence, the negative relationship between self-esteem (due to the fulfillment of psychological contracts) and Shame-NSE is clearer through the understanding that, in cultures that no longer rely on economic success, but self-realization (as a proxy of relational psychological contract dominance), “the self has to be flexible and fluid” (Fluck, 2000, p. 441). With the flexible self, the evaluative dimension of shame further decreases, as stated, “With … fluid self, shame can be experienced as a less threatening or disturbing emotion even though it stimulates critical thoughts on the whole self” (Seok, 2017, p. 141).

Conclusion

The perception of the ability of a company to fulfill a promise (or: to not disrupt the psychological contract) reduces proneness to a half (i.e. two of four) of moral emotions, i.e. the Guilt-REP (Repair) and Shame-NSE (Negative Self Evaluation). However, further analysis showed that the perception initially triggers an increase in Guilt-NBE (Negative Behavior Evaluation), which declined in the next phase, and subsequently rose again.

The findings are counterintuitive. Previous studies indicated that being not prone to moral emotion is capable of promoting corrupt behavior. Hence, based on the dynamic finding of the predictive correlation between disruption to psychological contract (PCB) and proneness to moral emotions, we can no longer assume that the fulfillment of promises by the company can suppress the tendency of corrupt behavior of bank employees.

That does not mean that the company does not need to fulfill its perceived promises—rather, the banking employee has unique characteristics to pay attention to, in terms of their moral emotion. For example, Lee and Gino (2019, p. 480) stated, “Indeed, individual's thinking about their identity as bank employees led to more cheating, which suggests that business culture may play a significant role in shaping one's moral compass.” The meaning is that, in order to support higher moral emotion proneness, the efforts of satisfying employees' psychological contracts need to be accompanied by a culture that can balance the calculative or quantitative mindset (Gino, 2014, para. 5)—that is omnipresent in the bank industry—with a more reflective mindset.

A managerial recommendation come from the results of this present study is that the company leaders need to assess the psychological contract breach/disruption on every employee in the banking industry. This assessment is useful for predicting when Guilt-NBE (Negative Behavior Evaluation) will reach a low point (thereby needing more intensive guilt inductions, e.g. through rule setting, clarifying responsibility, clarifying consequences, and increasing ‘victim-worth’; Gorta, 1998) based on certain breach levels. Practically, the assessment can be done through a computer-based or online psychological contract inventory. After obtaining a profile mapping of the perceived psychological contract breach, company leaders can discuss the maps with their employees in a learning or leadership forum. The learning forum can also bring together experts from the field of ethics or moral psychology to discuss improvement steps of the quality of fulfillment of psychological contracts as well as mitigation if there are some expected adverse effects from the identified breach.

For future studies, it is suggested that psychological contract breach inventory include additional domains besides the four existing dimensions (job satisfaction, salary and other compensation satisfaction, support, career opportunities), namely organizational policies, leadership, social contacts, as well as work-life balance (van Gilst et al., 2020). The four additional domains are very suitable for the context of Indonesia as a country with a collectivistic cultural orientation that emphasizes social relations and norms. Also, in measuring the moral emotions proneness, items that reflect the current situation (Covid-19 pandemic) could be added; for example, items asking the extent to which an employee feels ashamed and guilty when using his/her working from home (WFH) time for doing activities, both online and offline, not related to company missions and goals (i.e. organizational social loafing phenomenon). This suggestion is at the same time might present the connection between this present study and social reality nowadays.

Overestimated observed correlations between variables could be a limitation of this investigation because of the prevalent technique of collecting information from a single source, using a single technique at a single time (Juneman, 2013). Besides, the number of male participants which is, approximately, 1.75 times more than the number of female participants might make the generalization of the results of this study limited, e.g. more applicable to males.

Declarations

Author contribution statement

J. Abraham: Conceived and designed the experiments; Contributed reagents, materials, analysis tools or data; Analyzed and interpreted the data; Wrote the paper.

M. A. Kurniadi: Conceived and designed the experiments; Performed the experiments; Contributed reagents, materials, analysis tools or data; Wrote the paper.

E. W. Andangsari, M. M. Ali: Analyzed and interpreted the data.

R. H. Manurung, H. L. H. S. Warnars: Contributed reagents, materials, analysis tools or data.

Funding statement

This work was supported by The Research and Technology Transfer Office, Bina Nusantara University as a part of Bina Nusantara University's International Research Grant entitled “Corruption Prevention: The Roles of Media Literacy, Nostalgia, and Celebrity Worship” with contract number: No. 026/VR.RTT/IV/2020 and contract date: 6 April 2020.

Competing interest statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.