Due to the lasting effects of the economic crisis and our changing world of work, research on job insecurity is flourishing (Shoss, Reference Shoss2017). Job insecurity refers to the perceptions of employees that their jobs are at risk or that they are likely to lose their jobs (Schreurs, Van Emmerik, Günter, & Germeys, Reference Schreurs, Van Emmerik, Günter and Germeys2012). Based on theories predicting that job insecurity functions as a stressor or as an imbalance of the employment relationship, research has investigated its effects on a wide range of outcomes (Schumacher, Schreurs, Van Emmerik, & De Witte, Reference Schumacher, Schreurs, Van Emmerik and De Witte2016; Shoss, Reference Shoss2017; Sverke, Hellgren, & Naswall, Reference Sverke, Hellgren and Naswall2002; Yi & Wang, Reference Yi and Wang2015). For example, it is well established that job insecurity has negative effects on employee well-being, including mental and physical distress (Shoss, Reference Shoss2017; Vander Elst, Näswall, Bernhard-Oettel, De Witte, & Sverke, Reference Vander Elst, Näswall, Bernhard-Oettel, De Witte and Sverke2016), and may also have deleterious consequences for attitudes and performance (Cheng & Chan, Reference Cheng and Chan2008; Shoss, Reference Shoss2017; Yi & Wang, Reference Yi and Wang2015). Moreover, job insecurity may also have ramifications for career development and vocational choices (Klehe, Zikic, van Vianen, Koen, & Buyken, Reference Klehe, Zikic, van Vianen, Koen, Buyken, Perrewé, Halbesleben and Rosen2012; Zhao, Lim, & Teo, Reference Zhao, Lim and Teo2012).
Despite the convincing evidence that job insecurity is associated with numerous adverse consequences for employees, there is yet little research available focusing on the various ways through which people cope with job insecurity on a weekly basis. There is evidence that job insecurity fluctuates over time (including weeks: Schreurs et al., Reference Schreurs, Van Emmerik, Günter and Germeys2012), and it is relevant to investigate how weekly job insecurity relates to outcomes over the course of weeks. Moreover, while we know that job insecurity may negatively affect well-being, we know little about the ways individuals protect their well-being, and how they engage in reactive and proactive coping strategies to maintain their well-being in the face of job insecurity. In the contemporary neoliberal workplace, employees are increasingly expected to be active shapers of their own work as well as their environment in order to improve their own well-being (Grant & Ashford, Reference Grant and Ashford2008; Kooij, van Woerkom, Wilkenloh, Dorenbosch, & Denissen, Reference Kooij, van Woerkom, Wilkenloh, Dorenbosch and Denissen2017). Appraisal theory (Lazarus, Reference Lazarus1991) argues that depending upon people's appraisals, job security may be either perceived as a threat or a challenge. In the current study, we test the proposition that subordinates may actively shape their behaviors in response to perceived insecurity, thereby reducing the likelihood of suffering from poor well-being following job insecurity experiences. In line with the circumplex model of well-being (Warr, Reference Warr1994), we investigate subordinates' depression and enthusiasm as indicators of their well-being. We follow appraisal theory to differentiate between emotion-focused and problem-focused coping behaviors individuals adopt when responding to perceptions of job insecurity (Lazarus & Folkman, Reference Lazarus and Folkman1987). While individuals may cope with job insecurity through expressing their emotions or not (i.e., silence), they may also engage in problem-focused coping by either adapting themselves, or taking a proactive stance at work (see Griffin, Neal, & Parker, Reference Griffin, Neal and Parker2007; Klehe et al., Reference Klehe, Zikic, van Vianen, Koen, Buyken, Perrewé, Halbesleben and Rosen2012). Silence refers to withholding contributions to the organization beyond fulfilling one's core tasks, adaptivity is more reactive and represents efforts by employees to cope with the challenges imposed upon them by job insecurity. Finally, proactivity refers to employees initiating changes themselves in order to anticipate and counteract negative effects of job insecurity (Klehe et al., Reference Klehe, Zikic, van Vianen, Koen, Buyken, Perrewé, Halbesleben and Rosen2012). We expect that the use of emotion-focused (i.e., silence) strategies will result in more negative outcomes compared to adaptive and proactive approaches, ultimately leading to lower well-being at work (Richter, Näswall, De Cuyper, Sverke, De Witte, & Hellgren, Reference Richter, Näswall, De Cuyper, Sverke, De Witte and Hellgren2013).
However, it is unlikely that every employee will engage to the same extent in these behaviors. In line with a key tenet of appraisal theory that stress results from an interaction between situational and personal characteristics (Lazarus & Folkman, Reference Lazarus and Folkman1984), we introduce two key contextual factors: one at the level of the employee, and one at the level of the supervisor. First, at the employee-level, some individuals are better than others in regulating their emotions (Niven, Totterdell, Stride, & Holman, Reference Niven, Totterdell, Stride and Holman2011), and therefore are better able to direct their emotions in a more constructive way when they experience job insecurity (Fugate, Kinicki, & Prussia, Reference Fugate, Kinicki and Prussia2008). Second, at the supervisor-level, the supervisor will play an important role in addressing subordinates' experiences of job insecurity (Schreurs et al., Reference Schreurs, Van Emmerik, Günter and Germeys2012). We expect that in particular supervisory prosocial motivation is important, as supervisors who are highly prosocially motivated will be more likely to help others, especially employees who experience job insecurity (Grant, Reference Grant2008). While generally supportive supervisors may provide support to employees solely within their job and tasks, prosocial supervisors are also offering support beyond the job, and therefore, will also be more likely to provide support in the context of job insecurity as this may risk employees' continuation of their jobs, and thus may create a need to find employment elsewhere (Harrell & Simpson, Reference Harrell and Simpson2015). Therefore, prosocial supervisors may facilitate employees to direct their efforts such that these become most beneficial for the employee her/himself, and therefore improve employee well-being.
In sum, the research questions for this paper are: Does, and if so how, weekly job insecurity lead to poorer well-being? Do supervisory or personal resources mitigate against these relationships between job insecurity and well-being? We introduce a multilevel moderated mediation model (see Figure 1), in which subordinates' coping strategies mediate the relationships between job insecurity and well-being in the following weeks (all measured at the weekly level), and the relationships between job insecurity and employee behaviors are moderated by emotional regulation and supervisor prosocial motivation (contextual variables measured at trait level).
Conceptual model of the study.
The paper makes multiple contributions to the literature on job insecurity. First, this paper challenges conventional wisdom that job insecurity leads to lower well-being by showing that people may actively cope with job insecurity to protect well-being. People may engage in adaptive behaviors following job insecurity, and therefore retain enthusiasm and avoid depression at work. Furthermore, the paper shows that not all individuals are as adaptive, and that this depends on both the employee (i.e., one's skills to regulate emotions), and the context (i.e., whether one has a prosocial supervisor), thereby contributing to the question when is coping most effective (Folkman & Moskowitz, Reference Folkman and Moskowitz2004). Finally, we contribute by showing that the dynamics of job insecurity with well-being may differ substantially at the week level from the more general, cross-sectional, level. While cross-sectional research assumes that people are stable in their perceptions of job insecurity, the current study shows that job insecurity fluctuates over time (Schreurs et al., Reference Schreurs, Van Emmerik, Günter and Germeys2012), and that weekly perceptions of insecurity have spillover effects to following weeks, influencing how people behave and feel at work during a particular week. As cross-sectional research is unable to capture the weekly fluctuations and influences from weeks to following weeks, our study contributes to better understanding of the dynamic nature of job insecurity over time. Hence, our study may also enable managers and HR departments to develop and implement dynamic interventions before employees develop depression and lose enthusiasm in reaction to job insecurity.
Theory and hypotheses
Conceptual and methodological reasons to examine job insecurity over time
We study behavioral responses to job insecurity at a weekly level (Bakker & Bal, Reference Bakker and Bal2010; Schreurs et al., Reference Schreurs, Van Emmerik, Günter and Germeys2012) for various reasons. First, there is evidence that job insecurity fluctuates over the course of weeks (Schreurs et al., Reference Schreurs, Van Emmerik, Günter and Germeys2012), and therefore employee behaviors may actually fluctuate as well. Investigating behaviors following job insecurity using a weekly design allows for a proximal way of testing what employees actually do during weeks where they feel their jobs are at risk of loss. These perceptions may be dynamic, as external signals, such as communication from the organization or rumors, may vary over the course of weeks as well. Moreover, well-being is not a stable phenomenon, and over the course of time, work-related well-being is likely to change depending on what people experience and how they behave during preceding weeks (Bakker & Bal, Reference Bakker and Bal2010). Finally, job insecurity may affect well-being negatively in the long run, as prolonged exposure to insecurity may create long-term stress, and deplete people of their resources (Shoss, Reference Shoss2017). However, at the weekly level, job insecurity does not necessarily lead to negative effects, as people's abilities to cope with insecurity at a certain moment may prevent them from being affected in a negative way (Selenko, Mäkikangas, Mauno, & Kinnunen, Reference Selenko, Mäkikangas, Mauno and Kinnunen2013).
Job insecurity and its consequences for well-being
Research has argued and shown that job insecurity may function as an important job stressor (Vander Elst et al., Reference Vander Elst, Näswall, Bernhard-Oettel, De Witte and Sverke2016), as it indicates a perception of threat to one's work-related future. In the current study, two well-being indicators are relevant: work-related depression and enthusiasm (Madrid & Patterson, Reference Madrid and Patterson2014). These well-being indicators represent the low-activation, negative affect (i.e., depression) and the high-activation, positive affect (i.e., enthusiasm) dimensions of the circumplex model of affect (Warr, Reference Warr1994). On the one hand, depression refers to negative experienced affect at work and may result from exposure to job insecurity (Shoss, Reference Shoss2017), while on the other hand, enthusiasm represents enjoyment of and immersion in work. Enthusiasm and depression are usually negatively correlated, yet not mutually exclusive (Madrid & Patterson, Reference Madrid and Patterson2014). Previous research has shown that job insecurity is associated with high depression and low enthusiasm (Shoss, Reference Shoss2017). At the same time, recent research also shows that employees may be actively shaping their own work environment to improve well-being (Kooij et al., Reference Kooij, van Woerkom, Wilkenloh, Dorenbosch and Denissen2017; Parker, Williams, & Turner, Reference Parker, Williams and Turner2006). Hence, employees may also be motivated to act in ways to preserve their jobs (Selenko et al., Reference Selenko, Mäkikangas, Mauno and Kinnunen2013; Shoss, Reference Shoss2017).
Employees as active copers to their contextual environment
A key feature of job insecurity is uncertainty, which distinguishes it from other related constructs such as change or job loss (Shoss, Jiang, & Probst, Reference Shoss, Jiang and Probst2018). According to appraisal theory (Lazarus & Folkman, Reference Lazarus and Folkman1984), uncertainty of the event is the defining characteristic that leads to threat appraisals. In this regard, uncertainty about future events increases concerns regarding the extent to which individuals have control over events in their current environment (Lazarus, Reference Lazarus1991). Given the importance jobs play in one's life, job insecurity is likely to lead to stress (e.g., Bosman, Rothmann, & Buitendach, Reference Bosman, Rothmann and Buitendach2005; Shoss, Jiang, & Probst, Reference Shoss, Jiang and Probst2018) but it may also lead to mobilization of certain coping strategies in the individual to deal with the uncertainty of the future events and hence to take control of the current environment.
A core element of appraisal theory (Lazarus & Folkman, Reference Lazarus and Folkman1984) is the process of appraisal following an event which takes place in multiple steps: In primary appraisal, a person evaluates a particular situation with reference to the significance for his or her well-being. The following stage is secondary appraisal, in which an individual evaluates the harm; thus, in secondary appraisal, the coping options are evaluated. In the light of prior research (Stiglbauer, Selenko, Batinic, & Jodlbauer, Reference Stiglbauer, Selenko, Batinic and Jodlbauer2012) and context of our study, it is expected that individuals engage in coping strategies in response to job insecurity. These strategies include altering the situation, accepting the situation, escaping, or seeking social support (Hewett, Liefooghe, Visockaite, & Roongrerngsuke, Reference Hewett, Liefooghe, Visockaite and Roongrerngsuke2018). In explaining the consequences of job insecurity, we follow appraisal theory and propose that individuals are likely to engage in two broad ways of coping: emotion-focused and problem-focused coping. Emotion-focused coping is a passive coping strategy (Dewe, O'Driscoll, & Cooper, Reference Dewe, O'Driscoll and Cooper2010). It may involve ignoring the problem (known as ‘selective coping’), being silent or maintaining the status-quo (Dewe & Cooper, Reference Dewe and Cooper2007). Problem-focused coping, however, involves proactivity to reduce the stressor by addressing the problem and, as such, can be seen as an active approach to coping (Richman, Rospenda, Flaherty, & Freels, Reference Richman, Rospenda, Flaherty and Freels2001). Problem-focused coping can be done either reactively by changing oneself (i.e., adapt oneself), or proactively by changing the situation (i.e., proactively altering job tasks).
In light of these discussions, we utilize silence, adaptivity and proactivity as coping strategies employees are likely to adapt in response to job insecurity. Specifically, when employees experience job insecurity, they may engage in proactive behaviors, which are self-started and future-oriented behaviors that employees initiate to proactively counteract the possible negative effects that job insecurity might elicit. The three coping strategies range from withholding efforts to the organization (i.e., silence), to reactivity (i.e., adaptivity), to proactive behavior. They may be correlated with each other as individuals may engage in different behaviors during the same week to cope with job insecurity, but jointly explain the various coping strategies as defined by appraisal theory in response to stress (Lazarus & Folkman, Reference Lazarus and Folkman1984). Indeed, the dominant assumption and the bulk of the empirical evidence suggests that employees appraise job insecurity to be a hindrance stressor, i.e. a stressor that is regarded as detrimental to one's well-being rather than a challenging opportunity (Schreurs, Guenter, Jawahar, & De Cuyper, Reference Schreurs, Guenter, Jawahar and De Cuyper2015).
Given such an appraisal coupled with earlier empirical research on coping responses in reaction to perceived job insecurity (e.g., Berntson, Näswall, & Sverke, Reference Berntson, Näswall and Sverke2010; Schreurs et al., Reference Schreurs, Van Emmerik, Günter and Germeys2012, Reference Schreurs, Guenter, Jawahar and De Cuyper2015), we expect that in weeks where individuals experience job insecurity, they are generally likely to be more silent (see Schreurs et al., Reference Schreurs, Guenter, Jawahar and De Cuyper2015), less adaptive, and demonstrate less proactivity in the following weeks (see Mantler, Matejicek, Matheson, & Anisman, Reference Mantler, Matejicek, Matheson and Anisman2005; Stiglbauer & Batinic, Reference Stiglbauer and Batinic2015). However, a primary purpose of our study was to examine the conditions under which more positive coping strategies would be demonstrated by employees. Specifically, we propose that employees high in emotional regulation and who work under supervisors with a more prosocial orientation are more likely to utilize effective coping strategies (more voice and proactivity, less silence) compared to their counterparts who are low in emotional regulation or work with a supervisor low in prosocial orientation. Before discussing the rationale for these moderating hypotheses, we first discuss the well-being outcomes associated with employee selection of coping mechanisms to further justify the theoretical and practical importance of focusing on this topic.
Impact of coping strategies on well-being
Appraisal theory argues that the effectiveness of coping for well-being depends on the type of coping strategy (Dewe, O'Driscoll, & Cooper, Reference Dewe, O'Driscoll and Cooper2010). It is generally accepted that more active, problem-focused strategies are more effective than passive strategies for protecting individuals against negative outcomes (e.g., Dehue, Bolman, Völlink, & Pouwelse, Reference Dehue, Bolman, Völlink and Pouwelse2012). This is mainly because active strategies attempt to remove or control the stressor, whereas passive or emotion-focused strategies aim to modify the responses of the individuals to the stressor; rather than eliminating the source of stress (Folkman & Moskowitz, Reference Folkman and Moskowitz2004). In the light of the core tenet of appraisal theory, we therefore expect silence, a passive coping strategy, to be positively associated with depression and negatively with enthusiasm.
On the contrary, we expect active coping strategies, namely adaptivity and proactivity, to be more effective in dealing with job insecurity. Job insecurity may lead people to become more adaptive and proactive, both which aim to deal constructively with job insecurity, and thereby protecting their well-being. These strategies aim to address the core of the problem by trying to improve the situation (e.g., adaptive behaviors) or by taking pre-emptive actions to take control of the situation (Lechner, Bolman, & Van Dalen, Reference Lechner, Bolman and Van Dalen2006). Empirical research has supported this, revealing that health and well-being outcomes are usually improved as a result of adapting active coping strategies (e.g., Lee & Brotheridge, Reference Lee and Brotheridge2006). Moreover, proactive coping has been empirically associated with higher levels of engagement (Gan, Yang, Zhou, & Zhang, Reference Gan, Yang, Zhou and Zhang2007) as well as lower levels of depression (Greenglass, Fiksenbaum, & Eaton, Reference Greenglass, Fiksenbaum and Eaton2006). Taking all this together, we propose the relationships between job insecurity and well-being to be mediated by silence, adaptivity, and proactivity. In sum, Hypotheses 1 and 2 are:
Hypothesis 1: The relationship between job insecurity and depression is mediated by (a) silence, (b) adaptivity, and (c) proactivity.
Hypothesis 2: The relationship between job insecurity and enthusiasm is mediated by (a) silence, (b) adaptivity, and (c) proactivity.
To summarize, we expect that job insecurity will be related to higher levels of depression and reduced enthusiasm (two outcomes commonly seen in response to job insecurity) and that these relationships can be explained by the choice of coping strategies used by employees in response to that job insecurity. However, as shown in our conceptual model in Figure 1, we also propose a first-stage moderated mediation model, such that employee usage of these different coping strategies will vary as a function of two important contextual variables: employee emotional regulation and supervisor prosocial motivation.
Buffering effects of emotional regulation and supervisor prosocial motivation
In addition to expecting mediated effects of weekly job insecurity on well-being via employee behaviors, we also argue that these relationships are contextually determined. Appraisal theory points out that the likelihood of people to respond to a stressor in a particular way depends on the interaction between the personal characteristics of the focal person and the context (Lazarus, Reference Lazarus1991). Hence, it is likely that these evaluations depend on either the person her/himself, or on the available support from the environment, and in particular the supervisor. In other words, the extent to which employees select one coping strategy over another in response to the perceived threat of job insecurity is likely to differ on the basis of their capability to materialize appropriate behavioral responses as well as the support they receive to do so. We therefore introduce two key contextual factors, which are likely to moderate the relationships between perceived job insecurity and the selection of coping strategies.
Klehe et al. (Reference Klehe, Zikic, van Vianen, Koen, Buyken, Perrewé, Halbesleben and Rosen2012) argue that employee characteristics as well as situational characteristics will moderate the extent to which employees are likely to react to job insecurity with adaptive coping strategies. Specifically, they argue that more adaptive coping is likelier when employees receive adequate social support that ‘signals psychological safety’ needed for engaging in proactive behaviors and using one's voice (rather than remaining silent). Moreover, they contend that individual differences characteristics (e.g., personality, knowledge, beliefs) will affect these relationships as well. Therefore, in the current study, on the one hand, we argue that at the personal level, employees' emotional regulation style will affect the extent to which they have control over their behaviors, such that they are likely to respond more favorably to perceived job insecurity. On the other hand, employees are dependent on their supervisors in terms of receiving the necessary support to be able to respond constructively to job insecurity and thereby maintaining well-being. Thus, we specifically focused on employee-level emotional regulation and supervisor-level support, operationalized as prosocial motivation.
Emotional regulation refers to individuals' generalized tendency to react in a certain way to manage their emotions (Niven et al., Reference Niven, Totterdell, Stride and Holman2011). Ample research has shown that people have generalized strategies which they use to regulate their own emotions in daily life, and particularly in adverse situations (Gross & John, Reference Gross and John2003). Appraisal theory dictates that the extent to which a potential threat (such as job insecurity) is perceived as a stressor depends on primary appraisal and interpretation, which is likely to be affected by people's skills to regulate their own emotions. Hence, emotional regulation as a trait concept is likely to affect the extent to which people are interpreting job insecurity as a threat which needs to be counteracted by coping behaviors. Individuals with better emotional regulation may be more likely to appraise job insecurity as a challenge stressor, and one that can be managed via voice and proactivity, rather than a hindrance stressor that is best managed via increased silence, so as not to further jeopardize one's precarious position (i.e., loss aversion).
In the current study, we focus specifically on affect-worsening emotion regulation (Niven et al., Reference Niven, Totterdell, Stride and Holman2011), which is described as the tendency to deliberately exacerbate or intensify one's own feelings when experiencing negative events. We argue that affect-worsening emotional regulation is important in the context of job insecurity, as it may influence the selection of adaptive versus maladaptive coping strategies. Specifically, employees who are prone to ruminate and intensify negative thoughts following threats may be more prone to loss aversion and engage in silence so as not to further jeopardize their seemingly precarious position within their company.
Cynicism and ruminating on negative experiences, which are core elements of this emotional regulation strategy, tend to have negative effects on how people feel about themselves and their capability to overcome obstacles, and therefore, it is likely that employees high on affect-worsening emotional regulation are less able to interpret job insecurity positively, and therefore will engage in more silence and less adaptivity and proactivity. Thus, we expect that for those people high on affect-worsening regulation, the relationships between job insecurity and employee behaviors are accentuated. In contrast, people with low affect-worsening regulation will be better able to engage in voice (thus not to remain silent), adaptivity, and proactivity following job insecurity in a particular week. Hypothesis 3 therefore is:
Hypothesis 3: The relationships between job insecurity and silence, adaptivity, and proactivity are moderated by affect-worsening emotional regulation, such that individuals high in affect-worsening regulation will be more likely to choose silence and less likely to choose adaptivity or proactivity compared to individuals low in affect-worsening emotional regulation.
In addition, we expect supervisor prosocial motivation to moderate the relationships between job insecurity and employee behaviors. Appraisal theory proposes that it is not only individuals themselves who make decisions on whether or not to engage in coping, but the environment plays an important role as well (cf. Klehe et al., Reference Klehe, Zikic, van Vianen, Koen, Buyken, Perrewé, Halbesleben and Rosen2012). In the current study, we focus especially on supervisors, as they are close to employees, and have access to resources and have power to either help or hinder employees in their work-related goals. We focus on prosocial motivation of supervisors, as it is especially those supervisors who are prosocial, who may be likely to help employees who perceive job insecurity. Prosocial motivation is the ‘desire to expend effort to benefit others’ (Grant, Reference Grant2008: 49), and can be considered an important motivation in the work domain, as it allows people to exert efforts to help others in the workplace. This is important in the context of job insecurity for at least three reasons. First, prosocial supervisors may be more aware of perceptions of job insecurity among their subordinates, as they may communicate more frequently with their subordinates about how they are doing than supervisors who are not prosocial. Second, prosocial supervisors may be more likely to help employees with job insecurity, and even help their subordinates in making suggestions to employees to enhance their employability beyond their organization, as they care for the employee and not only for the organization. Finally, an employee who experiences a prosocially motivated supervisor will be more likely to experience support from the supervisor, not only for the employee him/herself, but also for others in the department.
As job insecurity may affect not only workers themselves but also others around them (Shoss, Reference Shoss2017), supervisors with high prosocial motivation will be likely to help the focal subordinate as well as coworkers in creating a work climate in which workers can express their insecurity as well as a climate of trust in which workers are stimulated to bring forward new ideas and to be proactive (Grant, Reference Grant2008). Hence, we expect that especially when employees experience their supervisor being prosocially motivated, they feel supported, and thus following job insecurity, will be less silent, more adaptive, and proactive. Hypothesis 4 is:
Hypothesis 4: The relationships between job insecurity and silence, adaptivity, and proactivity are moderated by supervisor prosocial motivation, such that individuals with highly prosocial supervisors will be less likely to choose silence and more likely to choose adaptivity or proactivity compared to individuals with supervisors low in prosocial motivation.
Combining the four hypotheses results in a multilevel moderated mediation model, as shown in Figure 1, we expect silence, adaptivity, and proactivity to mediate the relationships between job insecurity and depression and enthusiasm. Moreover, we expect these relationships to be moderated by affect-worsening emotional regulation and supervisor prosocial motivation, such that we expect weaker indirect effects of job insecurity on well-being under conditions of low affect-worsening emotional regulation, and high supervisor prosocial motivation. Hypotheses 5 and 6 are:
Hypothesis 5: The mediated relationships between job insecurity and depression and enthusiasm are moderated by affect-worsening emotional regulation with weaker relationships under condition of low mood-worsening emotional regulation.
Hypothesis 6: The mediated relationships between job insecurity and depression and enthusiasm are moderated by supervisor prosocial motivation, with weaker relationships under conditions of high supervisor prosocial motivation.
Methods
Participants and procedure
The study was conducted among 149 employees in various organizations in Chile. All participants were professional employees who attended an MBA program at one of the major universities in the country. Data were collected through paper-and-pencil surveys among the participants when they attended their activities at the university. This method gave participants the chance to participate easily in this study, disturbing the regular work activities as little as possible.
Research assistants at the university introduced the project to participants in groups of 15, explained the purpose and value of the study, and highlighted participants' rights to confidentiality and voluntary participation. In all surveys, a cover letter accompanying the questionnaire indicated that the survey was being conducted solely for scientific purposes. The research did not involve any form of deception or risk to the participants beyond that encountered in everyday life, and the official research ethics committee of the University of Santiago approved the study. A personal identification code was used to allow for linking data across time. In the beginning of the study, participants filled out the trait measures, and subsequently filled out weekly questionnaires on the Friday afternoon at the end of the working week for 5 weeks (Bakker & Bal, Reference Bakker and Bal2010).
Once, they answered the survey, participants returned these to the researcher's assistant. To ensure that the data were sufficiently representative of weekly variations, it was decided to exclude participants who had provided less than 3 weeks of data. The analyses were therefore conducted on the remaining 97 participants (i.e., 65% of the original sample) who filled out at least three consecutive weekly questionnaires in order to test longitudinal effects over the course of the weeks.
The final sample was 48% male, the average age was 31 years (SD = 6.28 years; range 22–49 years), and 98% had at least an undergraduate university degree. Among the sample, 57% had professional jobs, 19% administrative jobs, and 12% had executive jobs. Sixty-nine percent of the sample worked in the private sector, 27% in the public sector, 53% worked in the service industry, 7% in manufacturing, and 37% in other industries. Finally, occupations of participants were distributed as follows: business/management professional (41%), civil engineer (34.5% percent), and psychologists (24.5%).
Measures
Data were collected at both the person- and week-level. In line with previous research conceptualizing and assessing job insecurity at the week level (Schreurs et al., Reference Schreurs, Van Emmerik, Günter and Germeys2012), we designed a weekly diary study in which respondents rated the extent to which they felt insecure of their jobs at the weekly level, as well as their weekly behaviors and well-being. This way of measuring allows for a more direct assessment of the activities that employees undertake in response to insecurity, thereby reducing recall bias. In general, short scales were used to reduce the burden on respondents to participate in the research (all 1–5 Likert scales).
Job insecurity (mean α = .90) was measured weekly with the four-item scale from De Witte (Reference De Witte2005). Items were adapted such that they reflected the experiences of employees during the particular week they looked back upon. An example item is: ‘This week, I worried about losing my job’.
Silence (mean α = .88) was measured with the three-item scale from Detert and Treviño (Reference Detert and Treviño2010), assessing the extent to which employees withhold their ideas and thoughts in the workplace. An example item is: ‘This week, I withheld ideas from my boss for changing inefficient work policies’.
Adaptivity (mean α = .79) was measured using three items from Griffin, Neal, and Parker (Reference Griffin, Neal and Parker2007), an example being ‘This week, I coped with changes to the way I have to do my core tasks’.
Proactivity (mean α = .92) was measured with three items from Janssen (Reference Janssen2000). An example item is: ‘This week, I initiated better ways of doing my core tasks’.
Depression (mean α = .77) was measured with a three-item scale from Madrid and Patterson (Reference Madrid and Patterson2014) asking ‘During the last week, how often have you felt in your workplace…?’ ‘depressed’, ‘dejected’, and ‘despondent’. This measure aligns with the circumplex model of affect, and distinguishes four types, two of which are relevant in the current study: low-activated negative affect (i.e., depression) and high-activated positive affect (i.e., enthusiasm).
Enthusiasm (mean α = .76) was also measured using three items, with the same instruction as used for depression. The items were: ‘enthusiastic’, ‘joyful’, and ‘inspired’.
Person-level variables
Both person-level variables were measured in the first week, and referred to participant's general perceptions of their emotional regulation style and supervisory motivation. Affect-worsening emotional regulation (α = .80) was measured with the affect-worsening sub-scale (four items) from the Emotional Regulation of Others and Self (EROS) scale developed by Niven et al. (Reference Niven, Totterdell, Stride and Holman2011). The items measure the general tendency of people to cope in situations with cynicism or focus on negative thoughts. An example item is ‘I express cynicism to worsen my feelings’.
Perceptions of supervisors' prosocial motivation (α = .94) were measured with four adapted items from Grant's prosocial motivation scale (2008). Items were adapted such that they indicated employee perceptions of their supervisor's prosocial motivations at work. An example is ‘My supervisor cares about benefiting others through her/his work’.
Analyses
The study provided data at both the week and the individual level, and hence, multilevel analyses are needed to test the hypotheses. To be able to conduct longitudinal analyses, we estimated the effects of job insecurity T1 to the mediators at T2 (i.e., silence, adaptivity, and proactivity), to the well-being measures at T3 (i.e., enthusiasm and depression). If respondents participated in all 5 weeks of measurement, we were able to have three sequences for these individuals (T1-T2-T3, T2-T3-T4, and T3-T4-T5), and for those respondents with missing weeks, we took the available sequence of weeks. In total, this provided us with 161 sequences of weeks following each other among 97 respondents, on which the final analyses were based.
Hypotheses were tested using multilevel path analyses in MPlus 7.4 (Muthen & Muthen, Reference Muthén and Muthén1998–2015). We used 20,000 Bayesian bootstraps to estimate direct and indirect effects. We tested a range of models. First, we tested a model to assess the unmoderated indirect effects of job insecurity on the outcomes (i.e., enthusiasm and depression). Furthermore, we tested a model with the moderating effects of emotional regulation and one model with the moderating effects of supervisor prosocial motivation. The multilevel path analyses allowed for estimation of both indirect effects as well as conditional indirect effects of job insecurity on the outcomes taking into account the significance of the interactions. The coefficients in the various models can differ substantially as a result of different predictors at the week and person level included. Independent variables at the week-level were person-mean centered, and the moderators were grand-mean centered, before creating the interaction term. For significant interactions, we estimated slopes and conditional indirect effects for one standard deviation below and above the mean of the moderator. Table 1 shows the means, standard deviations, ICCs, and both within-person and between-person correlations among the study variables.
Means, standard deviations, reliabilities, and multilevel correlations of the study variables
NLevel 1 = 438, NLevel 2 = 149; ICC = intraClass coefficient; *p < .05, **p < .01. Range of weekly reliabilities reported along the diagonal for level 1 variables, and overall reliabilities for level 2 variables. Between-person correlations above the diagonal, within-person correlations below the diagonal.
Results
Hypothesis 1 proposed that the relationship between job insecurity and depression was mediated by silence, adaptivity, and proactivity, while Hypothesis 2 postulated mediated effects of silence, adaptivity, and proactivity in the relationships between job insecurity and enthusiasm. We first tested the nonmoderated mediated model to estimate the mediated effects without the interactions included. This model included T1 job insecurity, T2 mediators, and the outcomes at T3 (i.e., enthusiasm and depression). Table 2 shows the results of the multilevel path analyses of the main effects model. The Deviance Information Criterion (DIC) was 445.19. To test whether the current model fitted better than alternative models, we also estimated the fit for a reversed-causality model (with T1 well-being predicting coping behavior at T2, which subsequently predicted job insecurity at T3), which obtained a poorer fit (DIC = 472.86). Hence, it can be concluded that the model fitted better than a reversed-causality model.
Unstandardized coefficients for multilevel path models testing main and indirect effects
NLevel 1 = 161, NLevel 2 = 97; **p < .01, ***p < .001. 20,000 Bayesian bootstraps used for estimates.
T1 job insecurity was unrelated to T2 silence (b = .06, ns) or proactivity (b = .09, ns) in the following week, but was negatively related to subsequent adaptivity (b = −.17, p < .05). Table 2 further shows that silence at T2 was related to higher depression at T3 (b = .15, p < .05), and that T2 adaptivity was significantly related to decreased T3 enthusiasm in the following week (b = −.28, p < .01), while all other effects of the coping behaviors were not significantly related to well-being in the following week (bs ranging from −.04 to .13).
Bayesian bootstrapped indirect effects of job insecurity on well-being showed that only the relationship between T1 job insecurity and T3 enthusiasm was significant and positive via T2 adaptivity (b = .04, p < .05). The other indirect effects were nonsignificant. Hence, Hypothesis 1 is rejected and Hypothesis 2 is supported only for adaptivity, while rejected for silence and proactivity. The within R 2 ranged from .00 for silence, to .11 for enthusiasm, indicating that 11% of the variance in enthusiasm during a given week could be explained as a function of job insecurity and coping behaviors in the preceding weeks. However, as we predicted that the mediation effects would be dependent upon the moderators, we continued by including the moderators separately into the model.
Hypothesis 3 predicted a moderating effect of emotional regulation on the relationships between job insecurity and silence, adaptivity, and proactivity. Figure 2 shows the results of the multilevel path analyses with emotional regulation as moderator in the relationships. The within-level R 2 were: silence .01 (incremental ΔR 2 = .01, in comparison to mediation model), adaptivity .05 (incremental ΔR 2 = .02, ibid), proactivity .01 (ΔR 2 = .00), depression .10 (ΔR 2 = .00), and enthusiasm .17 (ΔR 2 = .06). Emotional regulation moderated the relationship between job insecurity and adaptivity in the following week (b = .70, p < .05). As can be seen in Figure 3, the relationship between job insecurity and adaptivity was negative for low affect-worsening emotional regulation (b = −.45, p < .001), and positive for high affect-worsening emotional regulation (b = .20, p < .001), rejecting Hypothesis 3 for adaptivity as outcome, as we found that people who are prone to use affect-worsening emotional regulation become more adaptive when facing job insecurity.
Results for multilevel path analyses of job insecurity in relation to coping behaviors and well-being, and the moderating effect of emotional regulation.
Moderating effect of affect-worsening emotional regulation on the relationship between job insecurity and adaptivity.
Hypothesis 4 predicted that supervisor prosocial motivation moderated the relationships of job insecurity with coping behaviors. Figure 4 shows the results of the multilevel path analyses for prosocial motivation as moderator. The within-level R 2 were the following: silence .01 (incremental ΔR 2 = .01, in comparison to mediation model), adaptivity .05 (incremental ΔR 2 = .02, ibid), proactivity .02 (ΔR 2 = .01), depression .12 (ΔR 2 = .01), and enthusiasm .18 (ΔR 2 = .07). The interaction was nonsignificant in relation to silence (b = −.08, ns), but significant in relation to adaptivity (b = −.18, p < .001), and proactivity (b = .14, p < .001). Figures 5 and 6 show the interaction patterns. Figure 5 shows that job insecurity was not significantly related to adaptivity when prosocial motivation was low (b = −.01, ns), but negative when prosocial motivation was high (b = −.37, p < .001). This partially supports H4 for adaptivity, as the figure shows that people especially adapt when their experience of job insecurity is low, and supervisor prosocial motivation is high. Figure 6 shows that the relationship of job insecurity with proactivity is nonsignificant when prosocial motivation is low (b = −.03, ns), but positive when prosocial motivation is high (b = .25, p < .01). This supports Hypothesis 4 for proactivity.
Results for multilevel path analyses of job insecurity in relation to coping behaviors and well-being, and the moderating effect of supervisor prosocial motivation.
Moderating effect of supervisor prosocial motivation on the relationship between job insecurity and adaptivity.
Moderating effect of supervisor prosocial motivation on the relationship between job insecurity and proactivity.
Hypothesis 5 predicted conditional indirect effects of job insecurity on depression and enthusiasm especially when emotional regulation was low. Table 3 shows the Bayesian bootstrapped conditional indirect effects, using 20,000 bootstraps. In line with the significant interaction effect of emotional regulation on the relationship between job insecurity and adaptivity in the following week, we estimated that the conditional indirect effect of job insecurity on depression was negative via adaptivity when emotional regulation was low (b = −.35, p < .05), and positive when emotional regulation was high (b = .38, p < .05). For enthusiasm, similar bootstrapped estimates were obtained, with a negative indirect effect of job insecurity on enthusiasm via adaptivity when emotional regulation was low (b = −.35, p < .05), and positive effect when emotional regulation was high (b = .38, p < .05). This partially supports Hypothesis 5 for adaptivity, and rejects the hypothesis for silence and proactivity.
Bayesian bootstrapping tests for conditional indirect effects
NLevel 1 = 161, NLevel 2 = 97; *p < .05, **p < .01, ***p < .001. 20,000 Bayesian bootstraps used for indirect effects.
Hypothesis 6 predicted conditional indirect effects of job insecurity on depression and enthusiasm especially when supervisor prosocial motivation was high. Table 3 shows the Bayesian bootstrapped conditional indirect effects for the two significant interaction effects (on adaptivity and proactivity). For adaptivity as mediator, we found that job insecurity was related to higher depression when prosocial motivation was low (b = .21, p < .05) and nonsignificant when prosocial motivation was high (b = −.15, ns). Furthermore, we found that job insecurity was positivity related to enthusiasm via adaptivity when prosocial motivation was low (b = .20, p < .05), and nonsignificant when prosocial motivation was high (b = −.16, ns). This partially supports Hypothesis 6 for adaptivity.
Finally, the conditional indirect effects of job insecurity on depression via proactivity were nonsignificant at both low levels (b = −.10, ns) and high levels (b = .19, ns) of prosocial motivation, rejecting H6 for proactivity in relation to depression. However, support for H6 was found in the positive indirect effect of job insecurity on enthusiasm when prosocial motivation was high (b = .12, p < .05), while the indirect was nonsignificant when prosocial motivation was low (b = −.16, ns).
Discussion
Given the recent economic downturn and pandemic, job insecurity has become a main concern for many employees, with implications for organizations as well (Pérez-Nebra et al., Reference Pérez-Nebra, Sklaveniti, Islam, Petrović, Pickett, Alija and Sanderson2021). The main aim of the current study was to investigate how employees respond to job insecurity at the weekly level. Building on appraisal theory of stress (Lazarus, Reference Lazarus1991; Lazarus & Folkman, Reference Lazarus and Folkman1984) and research on job insecurity, we tested the indirect relationship of job insecurity with depression and enthusiasm via three behavioral coping responses: silence, adaptivity, and proactivity. Moreover, we integrated the cross-level moderation of employees' affect-worsening emotional regulation and supervisors' prosocial motives on the association between job insecurity and behavioral responses.
Our findings revealed that the associations between job insecurity and behavioral responses are moderated by (a) subordinates' affect-worsening emotional regulation and (b) supervisors' prosocial motivation. Regarding the former, the findings showed that for employees with low affect-worsening emotional regulation, job insecurity was negatively related to adaptivity in the following week, and lower depression and enthusiasm in the subsequent week. In contrast, for people high on affect-worsening emotional regulation job insecurity was positively related to adaptivity, and consequently with depression and enthusiasm in the following weeks. These somewhat counterintuitive results can be explained on the nature of affect-worsening emotional regulation; for people prone to worsen their affects in context of threats, their cynicism may drive them to adapt themselves more in the face of job insecurity. As this poses the most viable strategy to survive at work, they retain their enthusiasm at work, but at the same may also become more depressed with the prospect of potentially having to find new employment. This is also supported in the notion that depression and enthusiasm are not conceptual opposites, but may exist jointly during a particular week, where people on the one hand may enjoy their work and be enthusiastic about it, while on the other hand they may be depressed due to the risk of losing one's job.
Given that high affect-worsening emotional regulation involves cynicism, rumination, and continuous focus on the negative aspects of a situation (Niven et al., Reference Niven, Totterdell, Stride and Holman2011), job insecurity increases negative emotions associated with the situation and triggers the focal person to be continuously aware of the situation (e.g., Hershcovis, Cameron, Gervais, & Bozeman, Reference Hershcovis, Cameron, Gervais and Bozeman2018). Furthermore, the uncertainty of the situation leads these types of employees to continuously think and ruminate about their future. This may further alleviate the negative thoughts and actions, being central to their emotional regulation strategy. Therefore, rather than reflecting their toxic emotions and behaviors which may be potentially harmful in a team environment (Bakker, Van Emmerik, & Euwema, Reference Bakker, Van Emmerik and Euwema2006), adapting to the situation in the face of job insecurity may be a more beneficial strategy to maintain well-being.
Employees with prosocial supervisors engaged less in adaptivity, but more in proactivity in the following week. This supports the notion that prosocial supervisors are helping employees to cope in a constructive way with job insecurity, and push them to engage in proactive behaviors rather than adapting themselves to survive in the workplace. These proactive behaviors (i.e., problem-focused coping) resulted in more enthusiasm in the following week for employees with highly prosocial supervisors, but we found no other significant conditional indirect effects for prosocial supervisors as moderator. Hence, there is only partial support for the well-being effect of prosocial supervisors following job insecurity. Indirect evidence for this finding can be found in research that integrates prosocial motives with well-being: observing helping behaviors in others has been associated with reduced depression (Greenfield & Marks, Reference Greenfield and Marks2004) and increased well-being (Penner, Dovidio, Piliavin, & Schroeder, Reference Penner, Dovidio, Piliavin and Schroeder2005). This is because helping or observing others helping lead to positive affective states and decrease depression. For example, Grant and Sonnentag (Reference Grant and Sonnentag2010) found that prosocial motives had a buffering role on the association between negative task evaluation (conceptualized as a stressor) and emotional exhaustion. A potential reason could be the crossover of negative experiences from supervisors (e.g., avoiding help and ignoring the needs of co-workers) to subordinates (not engaging in proactivity).
Finally, we found for those employees with supervisors low on prosocial motivation, that job insecurity was positively related to depression and enthusiasm via adaptivity. For these employees, having to adapt without being supported by their supervisor, job insecurity may lead to higher depression, while at the same also higher enthusiasm. This may be explained on the basis that adaptivity at work is not a voluntary choice for them, as it results from their perception of job insecurity. Hence, these employees feel forced to become more adaptive, as they lack the necessary prosocial support from their supervisors, which may enhance their feelings of control and enthusiasm, while at the same time, also depressing them when they realize they lack the support in their organization.
Theoretical implications
Our study has a number of implications for theory. With regards to research on job insecurity, our study can be considered one of the first steps to explore how and why employees deal with their experienced job insecurity in a dynamic fashion over the course of weeks (Shoss, Reference Shoss2017). By adopting a within-person design, our findings showed that the experiences and perceptions of job insecurity may fluctuate over weeks and this is important for practice to design dynamic interventions for the management of job insecure employees. Our adoption of appraisal theory bridges between job insecurity and stress research. In particular, by conceptualizing job insecurity as a stressor and relating it to well-being of employees through three coping strategies, we contribute to recent research which has adopted appraisal theory to understand stress and well-being associations (e.g., Hershcovis et al., Reference Hershcovis, Cameron, Gervais and Bozeman2018; Stiglbauer et al., Reference Stiglbauer, Selenko, Batinic and Jodlbauer2012). Our adoption of silence, adaptivity, and proactivity (Griffin, Neal, & Parker, Reference Griffin, Neal and Parker2007) as employees' coping mechanisms can thus be considered a step to contribute to existing research and advances on appraisal theory, which has been confined to mainly two coping mechanisms, i.e., avoidance and confrontation, to date. Hence, our study contributes to the literature on job insecurity by showing such perceptions fluctuate over time, and that they may be related to specific coping strategies that people employ to protect their well-being. Moreover, our study contributes to the job insecurity literature by showing that responses to job insecurity are a function of both individual and contextual variables that determine the well-being effects of job insecurity.
Similar to the findings of recent research (Hershcovis et al., Reference Hershcovis, Cameron, Gervais and Bozeman2018), our results revealed that coping strategies interact with emotional regulation styles and supervisors in relation to well-being. This is in line with the tenet of appraisal theory that the consequences of stress appraisal (i.e., coping strategies) depend on the interaction between the stress source and the situational as well as personal characteristics (Lazarus & Folkman, Reference Lazarus and Folkman1984). To unpack this assumption and contribute to research on appraisal theory, we integrated employees' affect-worsening emotional regulation as personal and supervisors' prosocial motivation as situational characteristics to examine how and when the consequences of experienced job insecurity unfold. Theory on job insecurity may thus advance understanding of coping strategies following insecurity, through postulating more specific ‘fits’ between employees' tendencies toward certain emotional regulation styles and the behaviors they are likely to show in response to job insecurity. While some of these prove to be helpful in maintaining or even improving well-being, others may be more detrimental, and lead to higher feelings of depression, which may ultimately have negative consequences for employees.
Practical implications
The findings of this research showed that context, namely employees' affect-worsening emotional regulation strategies and supervisors' prosocial motives, influenced the consequences of experienced job insecurity. Our study therefore provides organizations and managers with indications to develop effective intervention tools concerning the consequences of job insecurity. Specifically, employees may benefit from training interventions, workshops, individualized coaching, and on-line mentoring programs around how to effectively maintain and manage affective states when dealing with perceived job insecurity (Hülsheger, Alberts, Feinholdt, & Langer, Reference Hülsheger, Alberts, Feinholdt and Langer2013). Organizations and managers are advised to reduce job insecurity as it may negatively affect well-being of their workers. One primary way to achieve this is to offer permanent contracts to employees, and refrain from offering temporary and precarious job arrangements to workers. Moreover, in case job insecurity is difficult to avoid, it is important for organizations to offer concrete ways for employees to proactively address these perceptions, such as training employees to be employable and more flexible so they can find jobs elsewhere in case they lose their jobs. This may include training focused on concrete job skills, but also broader skills, such as communication, negotiation, and presentation skills, so employees can benefit from updated skills that enhance their prospects in the labor market.
Regarding the impact of prosocial motives, an implication is that organizations should pay particular attention to cultivating and developing a resourceful work environment encouraging support and care for others, particularly among supervisors (Salas & Cannon-Bowers, Reference Salas and Cannon-Bowers2001). Drawing on our findings, we suggest that organizations implement training interventions aimed at fostering helping behaviors and developing more prosocial leadership styles to manage employees' experiences of job insecurity. It is important that leaders realize that they act as representatives of their workers and therefore have a duty of care toward their subordinates, and not merely act as managerial tool to increase productivity of workers at the expense of their health and well-being.
Study limitations
Despite the strengths, the study has some limitations. First, the variables were collected using a self-report questionnaire. Therefore, we followed several recommendations to reduce method bias, as proposed by Podsakoff, MacKenzie, and Podsakoff (Reference Podsakoff, MacKenzie and Podsakoff2012). According to Evans (Reference Evans1985), method bias is less of a problem when testing moderation effects because estimates for the interaction effects remain accurate even if the data are collected from the same source. It can thus be concluded that method bias was less of a concern in our design.
Second, the study was conducted among a group of professional and well-educated Chilean professionals. They are likely to be better equipped with skills and educational levels to find new jobs if they experience job insecurity. Hence, the study findings cannot be merely generalized to a wider population of employees. We suggest future research to explore the job insecurity experiences among nonprofessionals who may face troubles managing the consequences of their job insecurity experiences.
This study was conducted in a unique cultural setting (e.g., Chile). Given that the cultural characteristics of this context emphasize high in-group collectivism and uncertainty avoidance (House, Hanges, Javidan, Dorfman, & Gupta, Reference House, Hanges, Javidan, Dorfman and Gupta2004), employees may refrain from reflecting on and reporting their job insecurity concerns. Future research might explore the extent to which corporate and national cultural characteristics influence the way job insecurity is experienced.
Conclusion
This study investigated weekly employee behavioral responses to job insecurity among a sample of Chilean professionals over the course of 5 weeks. The results showed that while job insecurity in general is unrelated to work behaviors and well-being at the weekly level, it also showed relationships were either positive or negative depending on employees' emotional regulation styles, and in particular the extent to which they engaged in mood-worsening regulation. For people high on mood-worsening regulation, adaptive behaviors were more beneficial for well-being. Moreover, supervisor prosocial motivation was also important in the context of weekly job insecurity to be either adaptive or proactive at work, and therefore to sustain one's well-being. Thus, weekly experienced job insecurity may affect behaviors and well-being dependent on emotional regulation and the supervisor. Our study finally showed that job insecurity may relate to behavior on the job in the subsequent week, and well-being in the week thereafter. Hence, our study shows that job insecurity has important dynamics across weeks in the workplace.
