Aiming at the Organizational Sustainable Development: Employees’ Pro-Social Rule Breaking as Response to High Performance Expectations

Despite the continuous increase in empirical research on pro-social rule breaking (PSRB), why organizational members conduct this behavior volitionally still requires further exploration. Drawing on the conservation of resources theory, our study investigated the impact of leaders’ high performance expectations on employees’ PSRB, following a hypothetical model with work stress as the mediator and perceived organizational support as the moderator. A three-waved time-lagged survey covering 208 dyad data of supervisor-subordinate from 41 teams of five enterprises in Shanghai, China, provided support for our hypotheses. After analyzing, we found that high performance expectations increased employees’ work stress, and further influenced employees’ PSRB substantially via stress, where the relationship was moderated by perceived organizational support. The theoretical and practical implications are discussed from a sustainability perspective.


Introduction
Organizational rules are considered essential to maintain organizational fairness and stability, improving organizational efficiency, and directing organizations toward sustainable development goals [1][2][3][4]. Regrettably, rigid rules prevent organizations from adapting to the rapid changes in the external environment, as there is a conflict between rigid organizational rules and flexible employee responses, which impairs the sustainable development of employees and organizations as well [5,6]. As a result, employees at the workplace frequently violate formal rules to accomplish what is good for the organization or others [7,8]. For example, the waiter will give the frustrated customer a free dessert to retrieve the situation and satisfy the customer. The behavior that employees intentionally violates a formal organizational policy, regulation, or prohibition with the primary intention of promoting the welfare of the organization or one of its stakeholders is defined as pro-social rule breaking [7]. The existing research on the antecedents of PSRB mainly included individual factors, such as risk propensity and conscientiousness, and organizational situational factors, such as leadership style and ethical climates [7,[9][10][11]. However, the potential influences of work characteristics on PSRB, which is a kind of vital important predictor of employee behavior, are rarely unexplored, except by Morrison (2006), who indicated the positive relationship between work autonomy and PSRB [7]. Due to the ambidextrous nature of PSRB, empirical studies, including mechanism and boundary conditions of why and when the employees engage in PSRB are necessary for the sake of avoiding the negative effect of violation, as well as to promote its positive effect.
Managers put forward rigid requirements on employees in a highly competitive atmosphere as they intend to improve organizational performance [12]. Leaders' high performance expectations have an important impact on employees' work attitudes and behavior [13]. High performance expectations (HPEs), as a crucial feature of work, which involve discrete or comprehensive requirements about the heavy workloads, high complexity, and time limit, are widely implemented in organizations aiming to achieve an upward spiraling cycle of high performance through sustainable motivation to encourage employees' sustainable high-performance output, but its effects have always been controversial. While the researchers have proposed that HPEs can ensure the sustained improvement of personal performance [14][15][16], the extant literature relatively ignores the negative impact of HPEs; for example, employees will face psychological burdens, such as negative selfevaluation and low self-satisfaction when they fail to achieve organizational goals [17,18]. Drawing on the conservation of resource (COR) theory, employees have to invest more resources (such as time, energy) to avoid the loss of resources. Therefore, employees who are facing high performance expectations will bear more stress in the work because they need to meet more stringent performance standards compared to employees facing lowperformance expectations [19][20][21]. In the studies of organizational behavior, work stress is an important variable that affects employees' attitudes and behaviors, often being considered to harm traditional performance, such as causing violations, sabotage, and other counter-productive work behaviors. However, Fay and Sonnentag (2002) argued that work stress could motivate employees to take positive behaviors [22]. Since then, scholars have conducted more studies on the positive consequences of stress, but the consistent conclusion has not been reached. To sum up, we speculate that, in order to alleviate the stress brought by HPEs and to avoid the possible losses caused by the failure not to reach the targets, employees might elaborate on unconventional actions to achieve the performance goals set by leaders to be conducive to the sustainable development of individuals and organizations.
Based on the conservation of resource theory, organizational support is a kind of valuable social resource that plays an important role in helping employees improve their performance and achieve organizational goals. Numerous studies have shown that perceived organizational support could alleviate employees' work stress perception by influencing their general emotional response to work [23][24][25], as a buffer against stress at work, and also have a positive impact on organizational citizenship behavior [23] and risk-taking behavior [26]. Employees who receive consistent support from the organization are more willing to be responsible for the development of the organization and implement positive subsequent actions to achieve its goals [27]. The powerful psychological resource due to organizational support has a licensing effect on employees under high stress, encouraging them to engage in activities that benefit organizations or colleagues, even if the premise is to break rules. Therefore, we infer that perceived organizational support could moderate the relationship between stress and employees' PSRB.
This study explores the mechanism and boundary conditions of how the leader's HPEs affect employees' PSRB and the main contributions of this paper are as follows. First, we attempt to advance research on the antecedents of PSRB by investigating the linkage between HPEs and PSRB. Our study adds to the limited literature by introducing HPEs, which is a work characteristic reflecting the leaders' willingness and being endowed to the employee by their supervisors, and it is undoubtedly a vital factor that affects the behavior of the employee [28]. We also challenge previous literature that reflects the mismatch between the ability and resources of employees and the high requirements of leaders can only bring negative consequences [29]; instead, we argue that PSRB might be a practicable countermeasure. Second, we contribute to PSRB literature by examining stress as the mechanisms linking HPEs to employee PSRB from the perspective of conservation of resources theory, while previous literature on the mechanism of PSRB mostly focuses on social cognition theory, social exchange theory or social information processing theory. Faced with the stress of high expectations, employees possess a reasonable cognitive explanation for the implementation of PSRB from the perspective of weighing the gains and losses of resources. Third, our study adds to the knowledge of the stress and PSRB Sustainability 2021, 13, 267 3 of 17 by uncovering boundary conditions (i.e., perceived organizational support). Through the exploration of boundary conditions of PSRB, this study could provide insight to organizations and managers to guide employees facing high performance expectations to engage in pro-active behaviors to promote the sustainable development of individuals and organizations as well.

Pro-Social Rule Breaking and COR Theory
Based on previous studies, Morrison (2006) further proposed the concept of pro-social rule breaking [7] drew from Brief and Motowidlo (1986)'s "pro-social behavior" (refers to the behaviors of individuals to improve the well-being of others, groups or organizations) [30], and divided it into three main types: (1) improve efficiency, (2) help colleagues or subordinates, and (3) serve customers. According to the views and discussions of relevant scholars, the main motivation of PSRB is to help the organization or its stakeholders, that is, to be pro-social, which is the key to distinguish PSRB from destructive deviant behaviors driven mainly or even completely by self-interest or revenge motives [7,9]. Morrison (2006) initially explored antecedents on PSRB behavior in her pioneering research [7]. Subsequently, many scholars explored the predictors, boundary conditions, and consequences of PSRB from the perspectives of individual traits and environmental characteristics, and found that employees' PSRB could be stirred up to varying degrees by some variables, such as risk-taking propensity [7], conscientiousness [9,31], psychological safety [32], work autonomy [7,33], colleague behaviors [9,34], supervisor-subordinate relationship quality [35], transformational leadership [36], ethical leadership [37,38], ethical climate [11,39], and so on. HPEs are not only an important work characteristic variable, but also reflect the sustainable expectations of leaders to employees. Therefore, its effect on employee behavior is well worth studying.
The conservation of resource theory (COR theory) proposed by Hobfoll (1989) is mainly used to describe the process of the interaction of resources between the individual and social environment [40], aiming at better revealing and explaining individual behaviors in stressful situations. According to this theory, resources are defined as "individual characteristics, conditions, energy and other things that make individuals feel valuable or the way to obtain them". Resources are divided into four categories: material resources (e.g., cars, housing), conditional resources (e.g., friends, marriage), personality traits (especially positive personality traits, such as self-efficacy and self-esteem), and energy resources (e.g., time, money, knowledge). The COR theory holds that individuals with more resources are less vulnerable to losing resources and more capable of acquiring sustainable growth of resources, and vice versa. Therefore, individuals are more sensitive to the loss of resources than to the acquisition of resources, which is, avoiding the spiral of loss as much as possible; if conditions permit, people will try to accumulate resources and cultivate a value-added spiral to resist the potential loss of resources in the future.
The HPEs are often reflected in more complicated work content and higher standards, which often exceed the existing ability or resources of employees. Therefore, on the one hand, employees have to sustainably invest new resources (such as overtime, learning new knowledge and skills, trying new methods, etc.) to achieve the performance goals, but in the face of the uncertainty that these resources could be lost; however, once the performance targets are not met, employees may lose the trust of their leaders and colleagues, opportunities for promotion, funds, or even lose their jobs, resulting in the loss of existing resources of them. To avoid negative consequences, employees may take unconventional measures to attain their goals and to protect their resources pursuing sustainable development. PSRB is inherently risky [7,9], the implementation of the behavior is also considered to be a decision-making process involving risk trade-offs [10,11]. Therefore, it may be feasible to explore the generation mechanism of PSRB from the perspective of resource conservation.

High Perofrmance Expectations and Stress
In the era of the knowledge economy, competition among enterprises is more and more fierce. Thus, managers usually set performance goals that exceed the average level of the industry, and these goals will eventually be decomposed to every department and employee of the enterprise [41,42]. HPEs are ambitious goals or standards proposed by leaders to employees for their future work performance [43]. In the existing literature, researchers generally believe that setting higher goals is an important way to improve personal performance [14][15][16], and they contend that high-performance goals can guide employees to pay attention and pour energy into activities related to goals, motivate employees sustainably and eventually enhance their perseverance in achieving goals [41,43]. Besides, high expectations are closely related to employee innovation and other discovery behaviors, as well as the use of working knowledge and strategies [44].
Many of the studies have documented the positive relationship between challenging goals and employees' efforts and performance at a higher level, but ignored that the achievement of performance goals must be on the premise that employees can complete [45]. Once unable to reach the objectives, employees are prone to form negative self-evaluation, lose leadership support, and even risk losing their jobs. According to the COR theory, compared to the acquisition of new resources, people are less likely to tolerate the loss of resources that are already owned. To avoid the loss of resources, employees have to invest more resources (such as time, energy, etc.), but in reality, not everyone has the corresponding ability and resources to meet HPEs most of the time; thus, performance goals make employees feel anxious for success, struggle in the invalid work strategy, and further generate evaluation stress and performance anxiety [41].
Stress refers to the physical and mental tension caused by an individual's awareness of the imbalance between "needs and abilities". Caplan and Jones (1975) suggested that work stress is caused by the threat due to working environment characteristics on individuals, and argued that some work characteristics were stressors [46], such as workload, work complexity, role conflict, role ambiguity, etc. For employees, HPEs inevitably mean heavy workload, high complexity, urgent time constraints, rivalry, and so on. These factors act alone or in combination to bring about employees stress. When the high-performance indicators are higher than the expectation of employees, they will undoubtedly cause greater and sustained psychological stress on employees.
Based on the above analysis, the following hypothesis is proposed.
Hypothesis 1: High performance expectations are positively associated with stress.

Stress and Pro-Social Rule Breaking
In the early literature on stress, scholars generally believed that stress would have a negative impact on individuals and organizations. A large number of studies have shown that excessive stress or poor stressors would cause adverse effects on personal, physical, and mental health as well as work performance [47], such as low work efficiency, high turnover rate, and decreased job satisfaction [48]. However, with the rise of positive stress research and the concept of positive stress in the workplace, researchers realized that stress also had positive effects. Many achievements have been made in studies of the relationship between stress and positive behaviors. Many studies have documented that moderate or benign stressors, such as time constraints and work complexity, can positively promote employees' positive behaviors, e.g., creative behaviors, job shaping, and taking charge [49][50][51][52][53][54].
Work stress is often manifested as the ultimate performance stress in the workplace. Performance stress refers to an adaptive response to the degree of achievement of organizational goals, which will impact the physiology, mentality, and behavior of participants. When performance indicators beyond the expected level of employees arise, they will cause stress to employees. According to the COR theory, leaders' identification, positive self-evaluation, and work remuneration are all valuable resources for employees. Once per-Sustainability 2021, 13, 267 5 of 17 formance expectations are not met, employees will face the risk of losing these resources, such as losing leadership support, reduced remuneration, and negative self-evaluation. Moreover, under HPEs, employees are often limited by insufficient resources, which makes it more difficult for them to pursue goals or manage behaviors, thus encouraging employees to take more positive behaviors to solve the dilemma [55]. In this context, employees are inclined to improve risk tolerance; in other words, finding ways to fulfill the HPEs of leaders and keeping their existing resources, they try to avoid the formation of a loss spiral that leads to sustained losses.
Facing the HPEs of leaders, employees will deliberate countermeasures to achieve goals [56]. Challenging goals indirectly affects employees' behaviors by motivating them to actively seek task-related coping strategies. Schweitzer et al. (2004) also indicated that employees often resorted to unethical or irregular means rather than just doing their best under the rules when facing difficult work objectives [18]. According to the COR theory, under the premise of protecting existing resources, people will try to accumulate resources and cultivate a value-added spiral to create and harvest more surplus resources to resist future risks. In the circumstance that only breaking the rules could meet the high performance expectations presented sustainably by leaders, PSRB (e.g., bypassing some approval processes to improve efficiency, or allowing customers to exceed their decision authority without permission) may probably incur temporary criticism or even institutional punishment from leaders or organization, but leaders' expectations to employees are formulated around improving organizational performance, employees' violation is for the sake of organizational interests and to be pro-social (PSRB's dimension based on the interests of the organization), the realization of the goals will benefit the organization or others. Thus, in the long run, PSRB will bring higher leadership evaluation, self-efficacy, and colleague support for employees to make them obtain more resources and sustainable development. However, if an employee engages in unethical practices (e.g., profiting at the expense of others, making false statements about performance, etc.), the day the matter came to light might have been the day the employee was to be fired and lose all of his resources. Therefore, to meet the HPEs, employees tend to implement PSRB behavior when they have to.
Based on the above analysis, the following hypothesis is proposed.

Hypothesis 2:
Stress is positively associated with pro-social rule breaking.

The Moderating Effect of Perceived Organizational Support
Eisenberger et al. (1986) proposed Perceived Organizational Support (POS) and refers to employees' perception of the extent to which an organization pays attention to their contribution to work and cares about their interests [57], including the organization's support for employees' work, the concern for employees' interests, the recognition of employees' value and the sustainability of support policies. The higher is the perception of organizational support, the higher are the employee's sense of responsibility and emotional attachment to the organization [58]. Active and sustained organizational support promotes employees' work autonomy and creativity [59,60], and reduces the constraints on employee behavior. Thus, employees may violate the existing rules of the organization to safeguard the interests of the organization or stakeholders [61].
Organizational support can relieve the stress of HPEs on employees. First of all, employees who perceive the abundant support of the organization do not think they are "fighting alone". While the leader (or organization) puts forward HPEs-organizational support, supervisor support and colleague support positively affect employees' psychological security [62], create a psychological experience and atmosphere of trust and safety for employees, which relieve the stress of employees to some extent [63]. Second, under high intensity and high demanding work stress, an organization can enhance the perceived support of employees through human resource practices, such as employee participation in decision-making, job rotation, training and development, promotion, compensation, etc., and finally, provide a supportive environment for employees [64]. In such a context, organizational support can provide opportunities for the realization of individual values, thereby reducing emotional exhaustion and preventing the generation of stress. According to the COR theory, organizational support is a kind of valuable social resources that can produce their own resources. Effective organizational support can strengthen employees' POS and psychological security, both of which are important psychological resources for individuals. All the above further increase the chances of getting a value-added spiral and reduce the possibility of falling into the spiral of losing for employees, so POS would affect employees' PSRB [10].
According to the principle of reciprocity in social exchange, leadership support will trigger employees' gratitude [60,65]. Mayer et al. (2007) found that high supervisorsubordinate relationship quality could promote employees to implement PSRB by perceived support from the supervisor [35]. When employees feel concerned, recognized, and supported by the organization, the binding force of institution in their mind weakens, while the motivation to meet high requirements for the interests of the organization increases. Additionally, they believe that they have the obligation to help the organization achieve its goals. In this process, employees will implement more intra-role and extra-role behaviors (such as PSRB). Employees with higher POS have greater sense of responsibility and obligation to help the organization or other stakeholders reach their goals. When the existing rules and regulations of the organization prevent employees from seeking interests for the organization and stakeholders, employees will choose to violate the established rules and take PSRB behavior that is beneficial to the organization to improve their work efficiency, or help colleagues to complete work tasks more efficiently, or meet customer needs.
Therefore, employees with high POS may implement more PSRB behaviors in highly intense and stressful work. Based on the above, we pose the following hypothesis: Hypothesis 3: Perceived organizational support moderates the positive effect of stress on pro-social rule breaking, such that this relationship is stronger when perceived organizational support is higher.
Based on the theoretical derivation of the above hypotheses, we further propose a moderated mediation model: HPEs will produce a perception of high stress on employees, and indirectly affect the PSRB of employees positively; in other words, the stress perceived by employees play a mediator between the two. However, the intensity of the mediation effect is moderated by POS-that is, under different levels of organizational support, the indirect effect is also different. High POS will increase the employee's PSRB in a stressful context. Therefore, combining the above hypotheses, this study proposes the following hypothesis: Hypothesis 4: Perceived organizational support moderates the indirect and positive effect of high performance expectations on employees' pro-social rule breaking through stress. Specifically, stress will mediate these relationships under the condition of perceived organizational support such that this indirect effect is stronger when employees perceive a higher level of organizational support.
Based on the above analyses, we provide a graphical depiction of the proposed models in Figure 1. Specifically, stress will mediate these relationships under the condition of perceived organizational support such that this indirect effect is stronger when employees perceive a higher level of organizational support.
Based on the above analyses, we provide a graphical depiction of the proposed models in Figure 1.

Sample and Procedure
We sent out questionnaires to the employees of five small and medium-sized enterprises in Shanghai, China, which was involved in electronic information, equipment machinery, new material research and development, new energy, etc. The sample consisted of 293 full-time employees from 53 work teams, who provided questionnaires through email and with face-to-face interaction. Before issuing the questionnaires, considering the features of PSRB, with the consent of five corporate executives, we focused on collecting

Sample and Procedure
We sent out questionnaires to the employees of five small and medium-sized enterprises in Shanghai, China, which was involved in electronic information, equipment machinery, new material research and development, new energy, etc. The sample consisted of 293 full-time employees from 53 work teams, who provided questionnaires through email and with face-to-face interaction. Before issuing the questionnaires, considering the features of PSRB, with the consent of five corporate executives, we focused on collecting data from employees in departments with a high degree of autonomy, such as sales, quality control, design, etc. To minimize the common method bias [66], we collected data in three periods, one month apart. For the first time, data from HPEs and control variables were collected. One month later, on the second time, we collected data of stress and POS for the second time. Finally, we asked employees to fill out the PSRB scale. Then, questionnaires with missing data and/or invalid responses were abandoned. The final sample size of questionnaires that successfully matched the two-time points was 208 from 41 teams, and the effective rate of questionnaire recovery was 71%. From the sample statistics, the distribution of subjects in different categories and levels has been found to be relatively uniform. Of the participants, 46.6% were male, 9.1% were under 25 years old, 85.1% were 26-40, and 5.8% were over 41 years old. In terms of education level, high school and below accounted for 26.4%, junior college 26.5%, and bachelor's degree and above were 47.1%.

Measures
All of the scales used in this study came from English literature. The translated items were revised by two academic experts majoring in human resource management, and the appropriateness of the items was evaluated. Before the formal survey, the questionnaire was tested and a pilot survey was conducted to adjust the readability of the questionnaire. All the items used a 5-point Likert-type scale (1 = strongly disagree; 5 = strongly agree; see Appendix A).

High Performance Expectations
We used the 3-item scale of Podsakoff et al. (1990) to measure HPEs [67]. We asked team leaders to evaluate their high performance expectations on employees. Sample items were "I shows my subordinates that I expects a lot from them.", "I ask for only the best performance for my subordinates.", and "I will not let my subordinates settle for second best.". The reliability coefficient (Cronbach's alpha) of this scale was 0.815.

Stress
To assess stress, we used the 13-item scale from Zhang et al. (2014) [68], which has been validated in the Chinese context. We asked employees to describe the extent to which they felt stress at work. Sample questions were "how complex is the work that must be done" and "I have to perform multitask in my assigned projects.". The reliability coefficient (Cronbach's alpha) for this scale was 0.803.

Perceived Organizational Support
POS was measured using the 6-item scale of Eisenberger et al. (1986) [57]. Sample items were "The organization strongly considers my goals and values" and "Help is available from the organization when I have a problem." The reliability coefficient (Cronbach's alpha) of this scale was 0.903.

Pro-Social Rule Breaking
We measured the PSRB of employees with the 13-item scale of Dahling et al. (2012) [9]. Sample items were "I break organizational rules or policies to do my job more efficiently", "I violate organizational policies to save the company time and money", and "I disobey com- pany regulations that result in inefficiency for the organization." The reliability coefficient (Cronbach's alpha) for this scale was 0.910.

Control Variables
Studies have indicated that differences in demographic variables such as gender, age, and education level can affect PSRB of employees [7,9]. To control the influence of employees' characteristics and corporate characteristics on employees' PSRB, we controlled variables like gender, age, educational level, and industry in our analyses.

Analysis Strategy
In our study, we adopted an unmeasured latent method factor and Harman's single factor test to investigate the common method biases. MPLUS 7.0 was used to conduct confirmatory factor analysis of variables, and SPSS 22.0 was used to carry out descriptive statistics and variable correlation analysis. When testing the hypotheses, in addition to the hierarchical regression analysis, we used the method recommended by Edwards and Lambert (2007) to estimate the unbiased confidence interval of conditional indirect effects [69], to verify the effect of moderated mediation in the model. The confidence interval was set to 95%, and the bootstrap self-sampling number was 5000.

Common Method Biases Test and Confirmatory Factor Analysis
Following the suggestion of Podsakoff et al. (2003) [66], we first used the unmeasured latent method factor to test the common method variance that may be caused by the single source data collected in the measures. A method factor was drawn into the original fourfactor model of HPEs, stress, POS, and PSRB to construct a five-factor model. After testing the fitting indexes of the new model, it was found that, compared to the four-factor model, fitting indexes of the five-factor model had increased a little; however, the increasing extent of CFI, TLI, and RMSEA was less than 0.02, and the change of them was not significant. The results showed that the common method variance in our study was not significant. Furtherly, we used Harman's single-factor test to verify the final score. The results showed that the first factor accounted for 28.39%, indicating that there was no general factor that could account for the majority of the covariance among the measures, which proved again that there was no significant common method variance in our measures.
To effectively verify the discriminant validity of the major variables, we used MPLUS 7.0 to conduct confirmatory factor analysis on the four variables in the model: HPEs, stress, POS, and PSRB. Bandalos (2002) argued that the inclusion of all measurement items in the original model would result in certain parameter estimation bias because the recommended parameters to sample size ratio would be exceeded [70]. Thus, referring to Rogers and Schmitt (2004) [71], before the formal confirmatory factor analysis, we used item parceling for PSRB, POS, and stress, which adopted a high load strategy. The analysis results are shown in Table 1. The hypothesized four-factor model (χ 2 = 142.51; df = 48; CFI = 0.925; RMSEA = 0.097; TLI = 0.896) was superior to the other three models, which indicated that the discriminant validity of the four constructs in this study was acceptable.  Table 2 shows the means, standard deviations, and correlations of studied variables. An examination of the zero-order correlations provided initial support for our hypotheses. As expected, HPEs were significantly correlated with stress (r = 0.371, p < 0.01), and stress was positively correlated with PSRB significantly (r = 0.548, p < 0.01), which provided initial support for our hypotheses. Further regression analysis would be performed.

Tests of Hypotheses
Since the variables of the model are all at the individual level, being consistent with the Dahling and Gutworth (2017) test method [72], we use hierarchical regression to test hypothesis 1, hypothesis 2, and hypothesis 3. Besides the outcome variable, the other variables are centralized. The regression results are shown in Table 3. As shown in model 2 of Table 3, HPEs are significantly and positively related to stress (B = 0.237, p < 0.001), as expected in hypothesis 1. Meanwhile, the results of model 4 indicated that stress had a positive and significant effect on PSRB (B = 0.673, p < 0.001). Thus, hypothesis 2 was supported. Furthermore, we tested the moderating effect of POS on the relationship between stress and PSRB proposed in hypothesis 3. From the results of model 6, it could be seen that the interaction of stress and POS after decentralization is positively and significantly related to PSRB (B = 0.200, p < 0.05), which indicated that the POS played a positive moderator role in the relationship between employees' stress and PSRB. Additionally, we drew the moderation effect in Figure 2 according to the method of Aiken et al. (1991) [73], which further showed that when POS was high, the positive effect of stress on PSRB was stronger. Furthermore, we used the Johnson-Neyman technique to identify the regions in the range of the moderator variable where the effect of the stress on the pro-social rule breaking is statistically significant and not significant [74][75][76]. Figure 3 indicates that the score of −1.759 on the perceived organizational support can be regarded as a point of transition between a statistically significant and a nonsignificant effect of stress on pro-social rule breaking. As shown in model 2 of Table 3, HPEs are significantly and positively related to stress (B = 0.237, p < 0.001), as expected in hypothesis 1. Meanwhile, the results of model 4 indicated that stress had a positive and significant effect on PSRB (B = 0.673, p < 0.001). Thus, hypothesis 2 was supported. Furthermore, we tested the moderating effect of POS on the relationship between stress and PSRB proposed in hypothesis 3. From the results of model 6, it could be seen that the interaction of stress and POS after decentralization is positively and significantly related to PSRB (B = 0.200, p < 0.05), which indicated that the POS played a positive moderator role in the relationship between employees' stress and PSRB. Additionally, we drew the moderation effect in Figure 2 according to the method of Aiken et al. (1991) [73], which further showed that when POS was high, the positive effect of stress on PSRB was stronger. Furthermore, we used the Johnson-Neyman technique to identify the regions in the range of the moderator variable where the effect of the stress on the pro-social rule breaking is statistically significant and not significant [74][75][76]. Figure 3 indicates that the score of −1.759 on the perceived organizational support can be regarded as a point of transition between a statistically significant and a nonsignificant effect of stress on pro-social rule breaking.  To test the moderated mediation model proposed by hypothesis 4, we adopted the method recommended by Edwards and Lambert (2007) to estimate the unbiased confidence interval of conditional indirect effects [69]. The confidence interval was set to 95%, and bootstrap self-sampling number was 5000. When perceived organizational support was higher (+1SD), the indirect effects of the path HPEs-stress-PSRB were stronger than the lower (−1SD). The difference between groups was significant (B = 0.0405, 95% confidence interval [0.0018, 0.0858], excluding 0), so hypothesis 4 was verified. To test the moderated mediation model proposed by hypothesis 4, we adopted the method recommended by Edwards and Lambert (2007) to estimate the unbiased confidence interval of conditional indirect effects [69]. The confidence interval was set to 95%, and bootstrap self-sampling number was 5000. When perceived organizational support was higher (+1SD), the indirect effects of the path HPEs-stress-PSRB were stronger than the lower (−1SD). The difference between groups was significant (B = 0.0405, 95% confidence interval [0.0018, 0.0858], excluding 0), so hypothesis 4 was verified.

Discussion
High performance expectations as an effective management measure have gained increasing attention in regards to the motivational effect. Our study closely focuses on the problem of how high performance expectations affect employees' pro-social rule breaking behaviors. Through 208 questionnaires from five small and mid-size manufacturing enterprises in Shanghai, China, and based on COR theory, we shed light on the impact mechanism of HPEs on employees' PSRB, and explore mediating effect of stress and moderating effect of perceived organizational support to the relationship between them. From the perspective of organizational management and employee career development, we also responded to the currently hot research topic about sustainability in society and the technology field [77,78]. To cope with the fierce competition, it is normal for leaders to set high goals for employees, and the resulting stress will have an important impact on employees' behavior. In a favorable atmosphere of organizational support, employees intend to adopt PSRB behaviors as a positive way to achieve goals and to eliminate stress, so that they attain a win-win situation for individuals and organizations. Some meaningful conclusions are obtained: (1) stress is positively related to employees' PSRB; (2) HPEs of leaders indirectly and positively affect employees' PSRB through stress; (3) POS moderates the indirect and positive effect of HPEs on employees' PSRB through stress, and specifically, stress will mediate these relationships under the condition of POS such that this indirect effect is stronger when employees perceive a higher level of organizational support. Based on the findings above, our study discusses how to influence the sustainable development of an organization from the three dimensions of the manager, rules and employee behavior. Our findings enlighten managers on how to mitigate the negative effects of management by objective and to effectively manage the PSRB of employees.

Theoretical Implications
Our research provides value to the current literature on organizational behavior and the implications for the sustained development of the organization. Firstly, the results demonstrate that HPEs have a significantly and indirectly positive impact on employees' PSRB, that is, leaders' HPEs on employees will give rise to employees' PSRB behaviors. While there have been studies on the impact mechanisms of employees' PSRB at the individual and organizational levels (such as risk-taking tendency, leadership style, supervisor-subordinate relationship quality, and ethical climate, etc.), rare attention has been paid to the impact of an important situational factor in the workplace: HPEs on employees' PSRB. For HPEs, previous studies have focused on its positive side of improving organizational and individual performance, but relatively ignored that it may lead to negative employee behaviors [79]. However, our study highlights that employees would violate rules when they felt that it was difficult to meet the high expectations of the leader. As a new concept in the field of positive organizational behavior, PSRB belongs to positive organizational deviance. While previous researches showed that there were some negative effects, such as affecting organizational stability and easily bringing about negative evaluation, we argue that the main motivation of PSRB is altruistic and has a positive intention. In the long run, PSRB not only helps employees to meet the expectations of high performance of leaders, but also improves the overall performance of the organization to prompt sustainable development. Based on the COR theory, this paper expands the research on antecedent variables of employees' PSRB. The research conclusion is consistent with Piccolo and Colquitt (2006) [80], which confirmed that the HPEs set by enterprises would play a significant role in promoting employees' extra-role behaviors.
The conclusions enrich the theoretical discussions on the organizational context factors influencing employees' PSRB, and cast a new light on the further researches on PSRB.
Secondly, this paper attempts to open the theory "black box" about the process mechanisms of HPEs impacting employees' PSRB, and discusses the mediation effect of stress in the course. As a psychological and physiological reaction, stress can sustainably affect individual behavior and work performance [81]. While empirical analysis of this study found that perceived stress played a mediating role between HPEs and employees' PSRB. Drawing on the COR theory, from the perspective of individual perception of the gain and loss of resources, it provides a reasonable explanation for the generation of PSRB. HPEs will compel employees to complete challenging work in a limited time. If they fail to achieve their goals, there is a risk of loss in their resources (leader trust, self-efficacy, promotion, etc.), which often brings greater psychological pressure to employees; to eliminate stress and protect resources, employees will take various possible steps to achieve those requirements, even by breaking the rules of the organization. However, even if an employee violates the rules, considering the possibility of obtaining more resources for the sustained development of oneself in the organization, the violation behavior taken by him/her will be mainly aimed at the overall interests and long-term development of others or the enterprise, therefore it is pro-social. At the same time, personal interests are also taken into account. Thus, under HPEs, higher work stress will lead to more PSRB. This study promotes the development of work stress research and further demonstrates the important role of work stress in predicting employee positive behavior, which makes up for the shortcomings of existing researches.
Finally, we introduce POS into the research framework of the relationship between HPEs and employees' PSRB, and investigate the effect of POS as the boundary condition in the stated interaction mechanisms. The results show that POS has a positive moderating effect on the relation of employees' work stress and PSRB, which indicates that, with high POS, an increase of the stress would give rise to more employees' PSRB behaviors. This has something in common with the research conclusions of some scholars. George and Brief (1992) argued that when employees faced higher requirements and greater stress in work [82], POS could make employees feel the material support or spiritual encouragement given by the organization so that employees would have the conscientiousness to care about the development of the organization and to choose to make pro-social behaviors for the benefit of the organization and other members. At the same time, it also confirms the viewpoint put forward by O'Driscoll and Randall (1999) that employees who are supported by the organization are more likely to feel that they are responsible for the development of the organization and they should help the organization to achieve its goals [27].
This paper advances and deepens the research on the influence mechanisms and boundary of PSRB, and further expands the studies on employees' stress and their behaviors in the context of the sustainable development of individuals and organization.

Practical Implications
Our study indicated that sustainable HPEs can make employees have a positive perception of stress, and lead to employees' PSRB. Moreover, POS can positively moderate the relationship between stress and PSRB. Overall, the results provide strategic insights into organizational management and human resource practices.
On the one hand, managers should treat employees' PSRB correctly and take advantage of its positive side. In conventional cognition, the violations of rules by employees are destructive, but the PSRB is just altruistic and has a positive intention. This kind of behavior make managers realize the positive properties of violations and the untimeliness of organizational rules, and provide a new management perspective for leaders to manage employees. That is, they should properly utilize and guide employees' PSRB instead of just banning them.
On the other hand, we found that there was a significant positive relationship between stress and employees' PSRB. In other words, under HPEs, employees' perceived work stress at an appropriate level can urge them to implement PSRB, which is conducive to the achievement of organizational goals and the common interests of organizational members. However, some literature with respect to work stress and performance also point out that the relationships between high-intensity work stress and employee behavior or performance are more complex rather than simple positive relationships [83]. When employees' perception of stress is too low or too high, there may be different effects on employee performance and behavior [84]. As a manager, when setting performance goals for individuals or teams, they should understand the truth that 'Too Much Water Drowned the Miller'. Therefore, managers must establish sustainable performance goals in the management process, rather than pursuing high short-term benefits only to ensure the sustainability of employees' high performance output and achieve the purpose of promoting organizational performance. Employees also need to carry out stress management, improving their ability to withstand stress, keeping stress at a moderate level, maintaining physical and mental health, and improving work performance at length.
Finally, according to the conclusion of this study, it can be seen that POS plays a positive role in stress and PSRB. The management of employees' PSRB can be realized by strengthening their perception of organizational support. In the context of high work stress, managers should give sustained care and support to employees and execute high-level human resource management practices to help employees better cope with the stress, make them work harder, and realize organizational goals and profits.

Limitations and Future Research
Pro-social rule breaking is gradually becoming a new hot spot in the organizational behavior field. In effect, although the findings of this paper supplement existing literature on HPEs and employees' PSRB, there are still some limitations due to limited research capabilities and resources.
First, although the data were collected from different time points in this study, and the test results showed that there were no significant common biases, the causal relationship between variables could not be fully identified. In the future, researchers could utilize the longitudinal research method and experimental method to better verify the causal relationship between HPEs and employees' PSRB.
Second, there are limitations in the sample selection. Limited to the resources and time of the questionnaire survey, the sample enterprises are only limited to China, lacking the background of a different culture, which inevitably results in the regional limitations of the data and may affect the promotion of the research conclusions. When generalizing our findings to other cultural backgrounds in future research, caution is recommended. Therefore, the conclusions obtained in this paper need to be verified in a larger sample and a wider area.
Finally, most of the existing literature focuses on predictive variables of PSRB including this study. In the future, empirical or experimental research methods can be adapted to study the influence of PSRB behaviors on organizational performance, employee sentiment, employee career satisfaction or leadership management, etc.