Impact of organizational structure and social support on pro-social rule breaking: A frontline perspective

Abstract Scholars over the past two decades have shed significant light on employee pro-social rule-breaking behaviors. Yet, the organizational and contextual variables and their influence on these behaviors are not properly considered. This paper focuses on frontline pro-social rule breaking and examines the influence of two elements of organizational environment; namely, organizational structure and social support, on the three types of pro-social rule breaking, using cross sectional survey data of frontline employees of two Indian-scheduled commercial banks. The results show that organizational structure and social support have differential influences on job-oriented, customer-oriented, and co-worker-oriented pro-social rule breaking. Researchers deliberated the impact of structural elements; namely, participation in decision-making, hierarchy of authority, job codification, rule enforcement, and perception of social support, thus identifying the drivers and inhibitors of the behaviors. Paper also offers guidelines for managing the three types of pro-social rule breaking via managerial practices and policy interventions.


PUBLIC INTEREST STATEMENT
Pro-social rule breaking happen when employees take the risk of deviating from formal organizational rules to meet the demands of customers, co-workers, and organization in general. We examined frontline pro-social rule breaking through an empirical study among frontliners in Indian banks and the results show that the organization structure, inadvertently influences employee engagement of pro-social rule breaking. Organizational structural variables such as participation in decision-making, freedom in job/ task-related decisions, specification in job/taskrelated rules, rule enforcement, and the support individuals receive from immediate supervisor, peers, and work group members influence employee pro-social rule breaking tendencies. This study is important since extant research mainly focussed on personality attributes and leadership leading to these behaviours, ignoring the impact of situational cues. We believe the findings presented in the paper will enhance the understanding about pro-social rule breaking and also provide guidelines for managing the behaviours.

Introduction
Organizational rules direct employee behaviors and actions towards organizational goals (Brower & Abolafia, 1997) and rule following is considered essential for organizational sustainability (Vardi & Weitz, 2016). Regrettably, rules hinder organizational effectiveness in some contexts as they fail to deliver desirable stakeholder outcomes (Vardaman et al., 2014). Rules by nature are inflexible, and as a result, employees sometimes deviate from formal rules to meet situational demands and to safeguard customer, co-worker, and organizational interests. Morrison (2006) introduced prosocial rule breaking (PSRB) to describe such responses. PSRB happens when employees deviate intentionally from rules for the benefit of the stakeholders. Many researchers suggest that PSRB is universal and needs to be appreciated in organizations (Morrison, 2006;Vardaman et al., 2014). However, managers with their limited understanding, attribute PSRB to employee dispositional influences rather than to situational factors and typically slot PSRB with destructive employee deviance (Dahling et al., 2012). Interestingly, organizational factors like structure and social support also can be anticipated to activate PSRB and therefore putting the blame squarely on employees alone is a flawed approach.
Extant research on PSRB has reported many antecedents connected to job factors; namely, job autonomy and task complexity (Kahari et al., 2017); and individual factors; namely, job meaning, risk-taking propensity, conscientiousness (Dahling et al., 2012;Morrison, 2006), spirituality, moral conviction (Asadullah et al., 2019) and citizenship behaviors (Liu et al., 2019). PSRB is also linked to situational factors; namely, transformational leadership (Youli et al., 2014), ethical leadership (Zhu et al., 2018;Zhong et al., 2019), organizational virtuousness (Zeng, 2018), bureaucracy (Fleming, 2019) and paternalistic leadership (Tu & Luo, 2020). However, the potential influences of organizational structure and social support on PSRB are unexplored. Organizational structure represents the distinctive and comparatively stable traits of an organization, in terms of decision-making, job autonomy, job codification, and rule monitoring and can be anticipated to influence the engagement of PSRB, though inadvertently. Likewise, the perception of social support (supervisors, peers, and work group) also can be argued to influence PSRB tendencies. Therefore, the purpose of the paper is to examine the influence of organizational variables (structure and social support) on employee attitude towards pro-social rule breaking. Frontline employees (FLEs) of two scheduled commercial banks in India were surveyed using a structured questionnaire and the data were analysed using hierarchical regression to test the influence of structure and social support on PSRB.
FLEs in the banking industry provide an interesting context to investigative PSRB for several reasons. First, FLEs are boundary spanners who perform tasks and activities that are directly related to customer satisfaction. They interact with the customers' on one-to-one basis and frequently are under pressure to deviate from rules (Chung & Schneider, 2002). FLEs extra-role behaviors are found to affect customer positive emotions than in-role behaviors (Zhao et al., 2018). Second, a customer satisfaction study of Indian banks reported that customers have low satisfaction on employee responsiveness and customer-oriented attention (Ali & Ratwani, 2017). If so, it will be beneficial to probe employee attitude on PSRB as researchers suggest the importance of PSRB on customer satisfaction and organizational wellbeing (Morrison, 2006;Mayer et al., 2007). Third, the Indian banking industry is experiencing intense competition owing to global developments in the banking liberalization process (Fujii et al., 2014). Banks have undergone considerable restructuring in the last two decades, which have left lasting changes in workplace attitudes and employee behaviors (Preshita & Pramod, 2014). The study explores employee attitude on PSRB in the current organizational settings and provides original insights on PSRB in an Indian service setting, enabling a theoretical extension to PSRB and frontline employee literature.
The paper is organized as follows. The first section describes the theoretical underpinnings, hypothesis development, and a range of evidence from organizational behavior and public administration literature. The second section discusses the study design. The third section reports the results of hypothesis testing. The fourth and final section discusses contributions to the literature and practice followed by limitations of the study and directions for future research.

Pro-social rule breaking
The study draws on the scholarship of organizational behavior and public administration related to rule breaking. Morrison (2006, p. 6) conceptualized PSRB as "other-focussed" rule breaking, in contrast to the traditional perspectives about rule breaking as opportunistic (Eisenhardt, 1989). It covers deliberate employee deviance from top-down rules, policies, regulations enforced by the organization, and not the deviance from emergent and informal norms that develop within social groups (Levine & Moreland, 2006). Employees engage in PSRB with three aims: to execute job duties efficiently, to help a co-worker, and to serve customers. Morrison (2006) named the three types of PSRBs as job-related PSRB (JPSRB), co-worker-oriented PSRB (CWPSRB), and customeroriented PSRB (CPSRB) in line with the aforesaid aims.

Organizational structure, social support and their link to PSRB
Organizational structure and social support are elements of organizational environment. Of the above two, organizational structure is relatively stable and has two features; namely, centralization and formalization (Jimenez, 2017). Centralization refers to the allocation of power in the organization and has two sub-constructs; participation in decision-making and hierarchy of authority (Hage & Aiken, 1967). Participation in decision-making indicates the extent to which employees are permitted to contribute to the decision-making process; whereas the hierarchy of authority refers to the degree of autonomy employees are permitted within their jobs (Hage & Aiken, 1967). Formalization refers to the degree to which rules and procedures are formally written and communicated through members of the organization (Pugh et al., 1969) and has two subconstructs; job codification (Allison et al., 2016;Kelly, 2017) and rule enforcement (Johari & Yahya, 2018). Employees experience no or limited flexibility in their job roles under high formalization, wherein low formalization offers job flexibility. Social support at work is the employee perception that their wellbeing is considered by the social environment (Chiaburu et al., 2010;Eisenberger et al., 2002). Interestingly, both structure and social support have the potential to influence the engagement of PSRB. Leveraging on the reactance theory (Brehm, 1993), situational strength theory (Mischel, 1977), and social exchange theory (Blau, 1964), the purpose of the study is to examine the influence of organizational structure and social support on PSRB.
There are several reasons why the structural dimensions are expected to influence PSRB. First, prior research exploring the constructs similar to PSRB with constructive and rule-breaking attributes has used structure and social support as predictive variables. For instance, DeHart-Davis, 2007, p. 892, stated that organizational structure influences "unbureaucratic personality", who help in the organizational mission and display a tendency to deviate from rules. PSRB and unbureaucratic personality share commonalities. Likewise, prior research on workplace rule breaking or deviance utilized predictor variables that potentially imply different aspects of an organization's structure, specifically, centralization (Bennett, 1998a;Lawrence & Robinson, 2007;Leblanc & Kelloway, 2002;Sims, 2010) and formalization (Bozeman & Feeney, 2011;DeHart-Davis, 2007;Hoffman, 2008;Pulich & Tourigny, 2004). Second, bureaucratic work environments, which entail high centralization and formalization structures, have been criticized for rule-breaking incidences (Raelin, 1994). On the contrary, organizational structures that are the opposite of bureaucracy (adhocracy) that highlight limited centralization are suggested to reduce workplace deviance (Nelson-Horchler, 1991) due to flexibility, autonomy, and open communication (Greenberg, 1997a(Greenberg, , 1997b. Formalization and rule consistency are expected to improve rule following in organizations (Borry et al., 2018) and since PSRB refers to deliberate rule breaking it is propositioned that formalization will discourage PSRB tendencies. Further, Pugh (1966, p. 239) argued that scholars must pay attention to the changing employee attitude on structural properties leading to unintended employee reactions such as rule breaking. Likewise, previous research state that social support discourages workplace deviance (Maynard-Moody et al., 2003;Peterson, 2002;Vardaman et al., 2014). Since prior studies implicitly support the notion that organizational structure and social support components affect PSRB, a conceptual framework (Figure 1) is developed and the study investigates the influence of centralization, formalization, and social support on PSRB attitude using the same.

Theory and hypotheses
The projected relationship between PSRB and the organizational structure (e.g., centralization and formalization) can be explained based on reactance theory and situational strength theory. Reactance theory argues that when employees do not have the freedom to control different aspects of their job or work environment, they will engage in "reactance" to gain power and control over their job and work environment (Brehm, 1966;Marasi et al., 2018). However, the reactance will soon disappear and the individual will experience motivational arousal to engage in actions or behavior to regain the freedom and to control the situation (Brehm, 1993;Brehm et al., 1983). High levels of the hierarchy of authority and low levels of participation in decision-making create centralized work environments and it is argued, based on the modified reactance theory that since the employees experience loss of control and autonomy in coping up with the situational challenges and stakeholder expectations under centralization they are motivated to engage in PSRB. It is expected that employees with high levels of centralization perception; that is, low participation in decision-making and high hierarchy of authority will engage in greater PSRB. Specifically, a negative relationship is expected between participation in decision-making and PSRB, while the hierarchy of authority is anticipated to have a positive relationship with PSRB. Employees resort to more PSRB with less autonomy in the workplace (high levels of hierarchy of authority). But, when the hierarchy of authority is low organizations enable employees to perceive a sense of control ("job autonomy") over job-related decision-making (Hackman & Oldham, 1976, p. 258). Job autonomy enables "employee behavior to persist in the presence of barriers like organizational policies" (Galbraith, 1983, p. 15). Employees are able to handle difficult and frustrating situations, when they possess power and autonomy in their jobs and they are less likely to manage or to deal with the situation by engaging in deviant behaviors (Browning, 2008). Thus, the following hypotheses are proposed: Hypothesis 1: There is a negative relationship between participation in decision making (centralization) and PSRB.
Hypothesis 2: There is a positive relationship between hierarchy of authority (centralization) and PSRB.

Figure 1. Conceptual framework.
Situational strength theory (Mischel, 1977) explains the influence of formalization on PSRB. Situational strength theory states that "strong" or structured work situations (highly formalized structure) will motivate employees to engage in work behaviors that are more in sync with organizational expectations (Meyer & Dalai, 2009;Osgood et al., 1996). Employees have clear guidelines for what is expected of them and are given few chances to display their own individual differences in structured work situations. However, in "weak" or unstructured work situations (less formalized structure) cause ambiguous behavior expectations and do not provide rewards or punishments for engaging in specific behaviors. In unstructured situations, employees have the discretion to choose a particular course of action such as PSRB, in the best interest of the organization (Ambrose et al., 2016;Zimmerman, 2001). Formalization has two sub-constructs: job codification and rule monitoring. Job codification (Allison et al., 2016;Kelly, 2017) is a measure of how specifically rules describe what the employees are to do, while rule monitoring, discussed in this paper as rule enforcement (Johari & Yahya, 2018) is a measure to what degree the rules are enforced in the organization. High job codification and strict rule enforcement ensures "strong" formalized work situations, thereby promotes rule following (DeHart-Davis, 2007). If so it is argued that job codification and rule enforcement restrain PSRB. Further, alienation theory posits that formalization regulates professional norms of autonomy and control; it alters the professional view of the job, preventing on-the-job expressions of employee potential like engaging in PSRB (Organ & Greene, 1981). Therefore, it is construed that employees abstain from PSRB in a highly formalized environment. Thus, the following hypotheses are presented: Hypothesis 3: There is a negative relationship between job codification (Formalization) and PSRB.
Hypothesis 4: There is a negative relationship between rule enforcement (Formalization) and PSRB Social support captures the general social environment in the workplace, wherein the supervisors, workgroup members, and peers play vital roles. Karasek and Theorell (1990) defined workplace social support as the "overall levels of helpful social interaction available on the job from coworkers and supervisors" (p. 69). Following this definition, the study conceptualizes workplace social support as a combined support that an individual receives from the supervisor and coworkers. Employees are socially embedded in the organizational networks (Granovetter, 1985) and interactions within this network and the resultant social support enable employees to accomplish their work efficiently (Spreitzer & Quinn, 1996). According to social exchange theory, employees can form social exchange relationships, with their immediate supervisor (Liden et al., 1997) and co-workers (Deckop et al., 1999;Flynn, 2003). These different relationships have suggestions for employee behavior. Specifically, because employees return the benefits they obtain, they are likely to equal goodwill toward the party with whom they have a social exchange relationship (Masterson et al., 2000). Employee behavior can be viewed as motivated by a desire to seek rewards and to avoid potential costs in social situations. Employees perceive social support as a job resource (Humphrey et al., 2007;Oldham & Hackman, 2010) and therefore, may choose not to engage in costly behaviors (damaging the social relationships and support) like bending organizational and group norms. Hence it is argued that perception of social support will have a negative impact on PSRB. Thus, the following hypothesis is presented: Hypothesis 5: There is a negative relationship between social support and PSRB.

Instruments
The participants of the survey-based study were FLEs (customer associates) of two scheduled commercial banks in India. The estimated sample size was 370 (Cochran, 1977; estimate of N > 10,000; 5% margin of error; 95% confidence interval). With organizational authorization, researchers invited 750 randomly chosen FLEs to participate in the online survey sent via intranet. All the responses (520) were directly returned to the researchers. The responses with missing values were screened out, thus yielding 468 usable responses (response rate 62.4%).
The data collection was undertaken using specially designed online forms, having two sections. The first section used validated measures for capturing attitude on PSRB and employee perceptions on organizational structural dimensions, namely, centralization and formalization and social support. The study used a 16-item scale (Aiken & Hage, 1968;Dewar et al., 1980), which measures individual perception on the degree of centralization and formalization and has four subscales: participation in decision-making, hierarchy of authority, job codification, and rule enforcement. Index on participation in decision-making was developed from four questions such as "How frequently do you participate in decisions on the adoption of new policies?" Respondents specified the frequency in a 5-point Likert scale ranging from "never" to "always." Individual perception indexes on the hierarchy of authority, job codification, and rule enforcement were developed by averaging their responses for the statements using 5-point Likert scales ranging from "strongly disagree" to "strongly agree." Social support was measured using six items (Spreitzer & Quinn, 1996), and the attitude on PSRB was captured using 13 items (Dahling et al., 2012). All the measures used 5-point Likert scales. The second section of the form was designed to record demographic information such as age, gender, education, and job experience. Age, gender, and work experience were measured categorically. Age was measured across four categories starting from younger than 30 to older than 50 with two in between intervals reflecting ten years each. Gender was categorized as male, female, and others. Job experience was coded across four categories with less than one year to more than six years, with two intervals of three years.

Participants
The sample included a mix of respondents who belonged to various age groups with 42.5% of the respondents in the age group younger than 30, 37% in the age group of 50 and above, 13.5% of the respondents in the age group 40-49 and the remaining in the group 30-39. Here, 50.43% of the respondents were male and the remaining female employees. Most of the respondents (62.2%) had more than six years of job experience as FLEs in the bank, while 28.1% of the respondents had more than one year of experience as FLEs. Respondents with less than one year experience were very few.

Analysis
The study aimed to investigate the predictive influence of organizational structural variables and social support on PSRBs. The data obtained from the survey were cleaned and then summarised. Exploratory factor analysis using principal component analysis with Promax rotation was used. Even though the study does not focus on demographic variables, its influence on PSRB, structure, and support variables was analyzed using ANOVA and t-tests to identify the confounding effect. Correlation analysis and hierarchical regression analysis were used to assess the nature of the association and predictive ability of the factors of structure and social support on the three study variables, namely job-oriented PSRB (JPSRB), co-worker-oriented PSRB (CWPSRB), and customeroriented PSRB (CPSRB). The analyses were conducted using SPSS 22.0 version.

Scale reliability and validity
The scales were checked for internal consistency using Cronbach's Alpha. As a rule of thumb, the acceptable level of Cronbach's Alpha required for a construct is 0.7 or higher. The internal consistency of the centralization scale was 0.76, which of formalization was 0.80, which of social support was 0.86 and that of PSRB was 0.91 in the current sample. The scales were retained as such.

Common method variance
Common method variance (CMV) is likely as data were collected from a single source. Harman's single-factor test was conducted using exploratory factor analysis to check whether the covariance between the variables is an artifact of single-source common method bias. The test using all the items of the five variables in our study revealed that the first emerging factor accounted only for 21.8% (<50%) of the explained variance in the items thereby suggesting that single-source common method bias is not a significant problem in the current study. Morrison (2006) identified three categories of PSRBs based on the objectives; to perform one's job efficiently, to help co-workers, and to provide excellent customer service. Factor analysis of PSRB data was conducted using principal component analysis and Promax rotation method. The result of Kaiser-Meyer-Olkin (KMO) of sample adequacy obtained was 0.73 (>0.5) indicating sample adequacy for factor analysis to proceed. The Bartlett's test for sphericity resulted in 8,501 (p < 0.0) indicating the suitability of the data for employing factor analysis. Factor structure of 12 items, obtained after deleting one item (Table 1) showed that there are three factors; namely, JPSRB, CPSRB, and CWPSRB. The factor structure yielded was similar to the extant studies on PSRB (Dahling et al., 2012;Morrison, 2006).
Likewise, data reduction with principal component analysis with Promax rotation of independent variables, namely, centralization, formalization, and social support yielded five factors. KMO measure was 0.57, which is acceptable. Factors were named based on the composition linked to theoretical footings. The five factors of the independent variables are shown in Table 2.

0.64
Ignore organizational rules to cut "the red tape" and be an effective.

0.81
Break up organizational policies to save the company time, materials and money.

0.83
When rules interfere with job responsibilities, I break those rules.

0.56
Break the rules if my co-workers need help with. 0.77 Pass up breaking the rules, even if it would help a coworker.

0.74
Do not break organizational rules, even if a colleague is in need.

0.77
When another employee needs my help; I refuse to comply with organizational policies to help him/her.

0.74
Break rules that stand in the way of good customer service.

0.69
Give good service to clients or customers by ignoring organizational policies that get in the way with my job.

0.94
Bend organizational rules so to best assist customers. 0.87 Refuse to violate organizational regulations to give a customer what they want.

0.78
Source: Exploratory factor analysis of primary data; Note: PSRB-Pro-social Rule Breaking.

Influence of demographic factors
ANOVA and t-tests were used to investigate the effect of demographic factors, namely, gender, age, and job experience. The Independent sample t-test, a parametric test compares the means of two independent groups to determine whether there is statistical evidence that the associated population means are significantly different. ANOVA is useful when there are more than two groups. The results of the study are presented in Table 3.
The results showed that the influence of gender was not statistically significant on the dependent variable. Age and work experience had a significant influence on PSRB attitude. Favorable attitude on PSRB was reported high for 30-39 age category (41.64) followed by 40-49 (40.51), then by people younger than 30 (31.22) and people belonging to age group 50 and above (28.86). More than young employees (younger than 30), employees belonging to 31-39 groups reported that they would engage in PSRB. Incidentally, experienced employees are in favor of PSRB (43.0) than the less experienced employees. As people gain experience, they develop professional mastery and take control of their job situations. Therefore, they take the risk of engaging PSRB. Age and experience are controlled in the inferential analysis as they show significant influence over the study variable. (1) Social support The six items of this factor gather employee perceptions on the work related backing individuals receive from immediate supervisor, peers, and members of their work group. Spreitzer & Quinn, 1996 2.Participative decision making The items of this factor elicit employee perception on how far organization involves them in resource allocation and policy/new program decision making. Pandey & Welch, 2005;Johanim, 2010 3.Heirachy of authority The factor includes the items that take the response from the employees regarding the degree of freedom in job/ task related decisions. Pandey & Welch, 2005;Johanim, 2010Kelly, 2017Allison et al., 2016 4. Job codification The items of this factor take the response from the employees regarding how far job/task related rules are specified in the organisation. Kelly, 2017;Allison et al., 2016 5.Rule enforcement The factor includes item that take the response from the employees regarding the latitude of behavior that is tolerated. In other words degree of enforcement of rules in the jobs.

Johari & Yahya, 2018
Source: Exploratory factor analysis of primary data

Impact of structural variables and social support on PSRB
Descriptive analysis of the primary data for this study yielded mean and standard deviation (Table 4). Correlation analysis was undertaken to identify the strength and direction (Pearson coefficient) of the association between variables (Table 4). Participation in decision-making shows significant positive associations with JPSRB (r = 0.47) and CPSRB (r = 0.56). Hierarchy of authority displays weak associations with JPSRB (r = 0.09) and CPSRB (r = 0.22). Job codification has positive associations with JPSRB (r = 0.27), CWPSRB (r = 0.49) and CPSRB (r = 0.22). Rule enforcement has a negative relationship (r = −0.28) with CPSRB upholding the role of rule enforcements towards customer-related rule compliance. The three types of rule breaking show a strong correlation with each other, indicating that an employee, who is engaging in one type may hold favorable disposition to other rule breakings also.
The central aim of the study is to identify the influence of structural variables and social support on PSRB. Hierarchical regression analyses are carried out to identify which of the five factors (Table 2) influence employee attitude on the three types of PSRB, namely, JPSRB, CPSRB, and CWPSRB. Control variables (age and job experience) are entered in step one, and the five factors in step two. The details are given in Table 5.
Model for employee attitude on JPSRB is significant (R 2 = 0.54, ΔR 2 = 0.38, F (10.457) = 54.37, p < 0.01). Hypothesis 1 (JPSRB) suggests a direct negative influence of participation in decisionmaking on JPSRB; however, analysis shows a positive influence (β = 0.51, t = 13.38, p < 0.01), thus not supporting the hypothesis. Hypothesis 2 posits a positive relationship between hierarchy of authority and JPSRB and the result (β = −0.32. t = −5.60, p < 0.01) does not support the same. Hypothesis 3 suggests a negative association of job codification, but the analysis shows a positive influence (β = 0.34. t = 9.30. p < 0.01), thus not supporting the hypothesis. Hypothesis 5 posits a negative influence of social support on JPSRB, and the result (β = −0.20, t = −5.26, p < 0.01) supports the same. Rule enforcement did not significantly predict JPSRB and therefore hypothesis 4 is not supported.
Participative decision-making, hierarchy of authority, job codification, rule enforcement, and social support had significant effects on the CPSRB engagement. These variables together could explain 48% of the variance (R 2 = 0.67, ΔR 2 = 0.48, F (10,457) = 92.43, P < 0.01) in CPSRB. Hypothesis 1 (CPSRB) suggests a direct negative influence of participation in decision-making on CPSRB; however, analysis shows a positive influence (β = 0.40, t = 12.33, p < 0.01), thus not supporting the hypothesis. Hypothesis 2 posits a positive relationship with the hierarchy of authority and the analysis (β = 0.14, t = 2.87, p < 0.01) supports the same. Hypothesis 3 suggests a negative association of job codification but the analysis shows a positive influence (β = 0.38, t = 11.99, p < 0.01), thus not supporting the hypothesis. The hypothesis 4 relating to the influence of rule enforcement on CPSRB posits a negative influence and the analysis (β = −0.40, t = −11.10, p < 0.01) supports the same. Hypothesis 5 suggests a negative influence of social support on CPRB, but the analysis (β = 0.13, t = −11.10, p < 0.01) does not support the same.
The analysis shows that social support (β = −0.22, t = −6.11, p < 0.01) and rule enforcement (β = −0.20, t = −4.91, p < 0.01) have significant negative effects on the CWPSRB engagement and therefore hypotheses 4 and 5 suggesting the negative associations are supported. Meanwhile, job codification (β = 0.57, t = 16.25, p < 0.01) increases the indulgence of CWPSRB and therefore hypothesis 3 (CWPSRB) suggesting a negative influence is not supported. These variables could explain 33% of the variance (R 2 = 0.57, ΔR 2 = 0.32, F (10.457) = 64.20, P < 0.01) of the study variable. Participation in decision-making and the hierarchy of authority did not affect CWPSRB and therefore hypotheses 1 and 2 of CWPSRB suggesting the associations are not supported.

Discussion
The study provides empirical support for the proposition that in response to organizational factors, FLEs do take the risk of breaking organizational rules. Contrary to the anticipated negative relation (on the basis of reactance theory), the results show that employee participation in decision-making encourages both JPSRB and CPSRB tendencies. Participation in decision-making indicates employee structural empowerment (Lawler, 1992) and enables employees to take control of their work situations in an effective manner. Since rules by nature are inflexible and incompatible to the stakeholder interests in certain situations, employee may take the risk of deviating from organizational norms (Zhang & Xiao, 2020). Moreover, employee workplace involvements and productive risk takings are usually facilitated by some forms of recognition and support (Klaas et al., 2012). Therefore, FLEs break the rules to meet customer expectations and job requirements for organizational benefit.
Under high hierarchy of authority FLEs show a tendency to engage in CPSRB, as hypothesised based on the reactance theory. The finding is also in agreement with the literature on customeroriented deviance (Leo & Russell-Bennett, 2014) which reported that empathetic FLEs strive to make a positive service experience for customers even at the expense of their organization. However, the result shows that hierarchy of authority inhibits JPSRB. High degree of hierarchy of authority entails supervisors to control deviance with sanctions and punishments. Threat of punishments considerably reduces employee willingness to engage in PSRB (Fleming, 2019). This is understandable in the case of JPSRB, as it may be difficult to justify the rule breaking to disciplinary authorities.
The finding related to job codification having a strong positive relationship with all three PSRBs was not expected, since hypotheses based on situational strength theory anticipated a negative relationship. However, the finding on the relationship makes sense. For instance, Pandey and Welch (2005) reported that high level of job codification generates the negative perception of "red tape" among employees of public sector setting and employees with positive work attitudes have superior ability to overcome the constraints of high job codification. Moreover, rules, protocols, procedures in excess, promote the need to engage in PSRB (Borry & Henderson, 2020;Dahling et al., 2012;Fleming, 2019;Morrison, 2006;Piatak & Mohr, 2019). In line with the situational strength theory, the finding of the study shows that high rule enforcement in the workplace significantly reduces CPSRB and CWPSRB, whereas rule enforcement does not (statistically not significant) influence JPSRB. Employees may choose to desist from behaviors that lead to sanctions and disciplinary actions, even if deviance helps customers and co-workers.
Social support significantly predicts all three PSRBs. High perception of social support encourages CPSRB which is contrary to the proposed negative linkage. Masterson's (2001) analysis of social exchange in organizations based on organizational support theory, highlights that co-worker support of service employees has an effect on positive customer treatment. Likewise, good quality leader-member relationship indicating supervisor support is shown to motivate customer-oriented behaviors (Cha & Borchgrevink, 2018). Extending on the aforesaid prior studies, the current study also reports that FLEs are willing even to rule breaking for customer benefit, if they perceive high social support. As anticipated, perception of social support discourages employees from engaging in JPSRB and CWPSRB. The concepts of social exchange (Blau, 1964) and the norms of reciprocity (Gouldner, 1960) indicate that employee experiences of support create feelings of obligation that serves to increase functional behavior (Wayne & Green, 1993). In the current context, employees reciprocate the social support by not engaging in JPSRB and CWPSRB. Moreover, Bryant et al. (2010) have pointed out that managerial CWPSRB will upset the justice perceptions in the workplace and therefore employees experiencing high social support desist from CWPSRB. Further, as supervisors and peers are reported to rate task performance lower for employees who engage in PSRB (Dahling et al., 2012), employees may not be willing to breaking the job related and coworker-related rules, even when the situation demands rule breaking.

Implications
PSRB transpires because of an employee's desire to use their initiative and do what he or she believes is needed to perform the job in an effective, responsible, and responsive manner (Morrison, 2006). Organizations need to challenge the current assumptions on employee pro-social rule breaking, and in fact, should be amenable to use PSRB for sustainable performance. The study contributes to the burgeoning literature on PSRB by examining the influence of organizational structure and social support on PSRB. The findings confirmed that organizational structural elements and social support predict moderate to high PSRB variability and the results are compatible to that reported to the prior studies which have used some of the similar antecedents (Borry & Henderson, 2020;Breslin & Wood, 2016;Fleming, 2019). In addition to substantiating the relevance of the seminal research on PSRB (Dahling et al., 2012;Morrison, 2006), the study offers two important insights on frontline PSRB. First, the results show that organizational structure, though not intentionally, stimulates or weakens the engagement of PSRB. Second, the influences of structural and social support variables on the three types of PSRB are different.
The ideas presented in the paper help managers to better understand the organizational factors influencing PSRB and thereby caution managers against, attributing PSRB solely on employee dispositional factors and using poor task performance ratings on account of PSRB. PSRB is a reflection of frontline perspectives on structure and the rule effectiveness for achieving the organizational goal and therefore should be considered crucial to the knowledgebase residing at the organization-public interface. Though top and middle managerial perspectives are valuable, they will yield only a partial picture of the realm (Walker & Enticott, 2004). Managers should follow and use the intentions behind PSRB for rectifying the process delays and system redundancies. In addition, the principal findings of the study offer guidelines for organizations on the PSRB management through (a) empowerment mechanisms, (b) the job codification and enforcement, (c) job autonomy, and (d) social support. JPSRB can be supported through participation in decisionmaking, and can be discouraged through increasing job flexibility (installing minimal and essential rules), applying changes in the reporting structure to enhance job autonomy and by enhancing social support. Employees increasingly engage in CPSRB with participation in decision-making, hierarchy of authority, job codification, and social support. However, CPSRB can be controlled effectively with strict rule enforcement. CWPSRB can be actively discouraged through rule enforcement and social support.
The results of this study also contribute to the research on FLEs' discretionary extra-role behaviors. FLEs are customer-contact employees who perform their duties under the restrictions of both internal and external environments (Edmondson & Boyer, 2013), and their discretionary behaviors like CPSRBs influence customer feedback and customer outcomes. The results show that high level of codification facilitates CPSRB, whereas close monitoring restrains CPSRBs. Managers should identify the rules that FLEs must follow and monitor those rules closely to avoid rule breaking, whereas should permit autonomy around rules which are creating barriers to the achievement of organizational objectives.

Limitations and future research
This study has a few limitations. First, as the sample is collected from two Indian banks only, restraint should be applied whenever the findings are generalized to an organizational setting. However, as scholarship on deviance appeals for specific organizational or occupational studies for evolving knowledge of discretionary behaviors (Bowling & Gruys, 2010), organizational and occupational level-specific studies are recommended for the future. Second, single-source measures are prone to response artifacts such as social desirability bias and consistency effects that create spuriously high inter-correlations (Podsakoff et al., 2003). Though the current study is not significantly influenced by the common method bias, it is recommended that future research should use multiple sources such as supervisory and co-worker ratings for structural dimensions, social support, and PSRB ratings to validate self-reported ratings. Third, customer-related and jobrelated PSRB may possess potential implications for enriching customer encounters. But, organizations create and enforce rules to ensure standardization and discipline. So there is a possibility that PSRB, though hypothesized, as functional may turn dysfunctional. Hence, future research should focus attention on establishing the organizational consequences of PSRBs. Fourth, the study used social support as composite construct; however, role (main and moderating) of distinctive dimensions of social support based on the sources, nature (Cropanzano & Mitchell, 2005) needs to be pursued to clarify the implications.

Conclusion
PSRB can increase firms' adaptability and flexibility to unanticipated challenges and situations and provide a cue of eliminating or altering redundant rules. The study provides evidence of FLE attitude towards PSRB, as influenced by their perceptions of organizational structural features and therefore, PSRB can be managed through appropriate interventions. An examination on rules that are repeatedly broken may help the firm to identify the bottlenecks that impede customer service, job, and organizational efficiency.