key: cord-0824121-9p4k1uk4
authors: Lee, Sang-Hoon; Hur, Won-Moo; Shin, Yuhyung
title: Struggling to Stay Engaged During Adversity: A Daily Investigation of Frontline Service Employees’ Job Insecurity and the Moderating Role of Ethical Leader Behavior
date: 2022-05-18
journal: J Bus Ethics
DOI: 10.1007/s10551-022-05140-y
sha: c543cb4703e2db992b5ee38182ae20226574ce28
doc_id: 824121
cord_uid: 9p4k1uk4

Drawing on conservation of resources theory, this study examined the moderating role of ethical leader behavior in the effects of daily perceived job insecurity on work outcomes the next day (i.e., work engagement and customer-directed helping) through occupational regret the next morning among frontline service employees working in adverse work situations (i.e., the coronavirus disease pandemic). Using experience sampling method, data were collected from 135 frontline service employees across five consecutive workdays. The results showed that daily perceived job insecurity had a negative indirect effect on work engagement and customer-directed helping the next day through (increased) occupational regret the next day in the morning. In addition, ethical leader behavior moderated the negative indirect effect of daily perceived job insecurity on next-day work engagement and customer-directed helping through next-morning occupational regret. Specifically, these negative effects were especially stronger among employees who had observed low levels of ethical leader behavior the previous day. The theoretical implications of the present findings for researchers and their practical implications for managers are discussed.

The ongoing adverse circumstances surrounding the coronavirus disease pandemic have had widespread effects on frontline service employees, who represent their organizations during face-to-face customer service encounters (Liao & Chuang, 2004) . Especially of concern to many frontline employees is that they have been experiencing a lingering sense of job insecurity, which is defined as "an overall concern about the continued existence of one's job in future" (Sverke et al., 2002, p. 243 ). This may be partially attributed to the widely accepted notion that the service industry comprises a high proportion of jobs that may be vulnerable to uncertainty (approximately 65% to 75%). Also, employees in the retail, financial/insurance, and restaurant sectors are more likely to be susceptible to external disturbances (e.g., the COVID-19 pandemic) than other industries . Relatedly, concerns over the pandemic (e.g., social distancing and vaccination policies) has had an adverse impact on businesses that involve face-to-face interactions with customers on a daily basis (Koo et al., 2020; Oh & Lee, 2020) .

The notions above point to a timely issue that involves investigating the harmful effects of perceived job insecurity on frontline employees working in an unstable and unfavorable context (i.e., the COVID-19 pandemic and economic turmoil). Indeed, the pandemic, for example, has rendered the work contexts of frontline employees more volatile, hence making them witness daily fluctuations in their environments (i.e., confirmed COVID-19 cases and deaths, social distancing policies, vaccination policies). This is because frontline employees may perceive an increased likelihood of job loss due to escalating perceptions that they can be easily replaced (Ghilarducci, 2020) . As a result, these employees may feel that their future options and opportunities are limited on a given day, thus eliciting negative feelings and ultimately worsen service outcomes on subsequent workdays (Li et al., 2015) . Considering this uncertain context, disentangling employees' day-to-day emotional and behavioral reactions from an intraindividual perspective is important, as such volatile environments may have profound circumstances on the daily functioning of individuals (Garrido Vásquez et al., 2019) . However, while a growing body of research has explored the link between employees' daily perceptions of job insecurity and their daily work outcomes (e.g., Garrido Vásquez et al., 2019; Schreurs et al., 2012; Schumacher et al., 2021) , we still lack knowledge on why frontline service employees' daily experience of job insecurity under adverse working conditions may have detrimental effects on work outcomes and when these negative effects are mitigated or strengthened.

Drawing on conservation of resources (COR) theory (Hobfoll et al., 2018) and a within-person perspective on job insecurity, we argue that job insecurity is positively related to occupational regret on a day-to-day basis. Occupational regret reflects employees' negative evaluations of stressful situations and the exhibition of strong emotional reactions (Zeelenberg et al., 1998) . As such, it propels individuals to enter a defensive mode to preserve the self (Hobfoll et al., 2018) while being indicative of how an individual may anticipate one's perceived threat (Jiang & Lavaysse, 2018) . Based on these insights, we propose that occupational regret may help explain why perceived job insecurity may be negatively associated with crucial daily work outcomes (i.e., work engagement and customer-directed helping), as emotional reactions to job insecurity have been suggested to be proximal drivers of employee work outcomes (Jiang & Lavaysse, 2018) . As a result of assessing job insecurity and feeling negative emotions, employees are likely to engage in defensive coping strategies and detach from tasks. Thus, the first objective of our research is to assess the mediating relationship between job insecurity, occupational regret, and work outcomes on a daily level.

We further aim to explore the boundary condition that mitigates (or strengthens) the negative effect of daily job insecurity on daily work outcomes. Amid employees' efforts to remain engaged at work (i.e., work engagement) and their reallocation of limited resources to their expected tasks (e.g., customer-directed helping), leaders 1 may provide behavioral cues and psychological resources to sustain such work efforts (Fehr et al., 2015) . Indeed, prior studies have demonstrated the resource-replenishing role of diverse forms of leader behavior (e.g., servant or supportive leadership; Schreurs et al., 2012; Wu et al., 2020) in stressful situations by providing social and emotional resources to their followers. However, we further claim that ethical leader behavior may be particularly important for employees experiencing job insecurity under extreme conditions. We posit that these leaders not only benefit employees by replenishing their tasks and social resources but also, importantly, promote appropriate work conduct as leaders by exhibiting fairness and transparency in decision-making processes (Bormann, 2017; Lin et al., 2016) . Specifically, when leaders behave in an ethical manner, employees may perceive organizational policies and procedures as fair and credible, which alleviates uncertainty about their jobs (Loi et al., 2012) . Moreover, ethical leaders are likely to serve as credible sources of information that enhance the accountability and predictability of organizational decisions, thereby enabling employees to cope with uncertainty about their job (Loi et al., 2012; Quade et al., 2019) . Ethical leader behavior may also reduce employees' negative emotions and attitudes with substantial task and emotional resources (e.g., providing employees with care for personal hardships and with equal opportunities), which may facilitate meaningful workplace contributions (Lin et al., 2016) . Indeed, past job insecurity research has shown a moderating effect of ethical leadership on the job insecurity-work outcome relationship Loi et al., 2012) . Extending this line of research to the daily level, we contend that daily ethical leader behavior will alleviate (or aggravate) negative job perceptions, hence aiding employees to stay engaged (or disengaged) in their primary work activities (i.e., work engagement) and promote (or hamper) customer-directed helping. Thus, our second objective is to examine the moderating effect of daily ethical leader behavior on the daily relationship between job insecurity, occupational regret, and work outcomes.

Our research is expected to contribute to the literature in three key ways. First, we investigate the negative effect of perceived job insecurity from a within-person perspective (e.g., Schreurs et al., 2012) among frontline employees who have experienced an adverse context (i.e., the COVID-19 pandemic). Specifically, we examine how perceived job insecurity on a given day is associated with increased occupational regret the next morning, and ultimately worsens employees' work outcomes (i.e., work engagement and customer-directed helping) the next day. This offers practitioners knowledge for managing employees' job insecurity perceptions and resultant work outcomes on a day-to-day basis. Second, we advance the job insecurity literature by identifying a relatively under researched mediator of the relationship between job insecurity and employee outcomes, namely, occupational regret. We illustrate how a strong emotional reaction (i.e., occupational regret)-characterized by defensive withdrawal-derived from job insecurity the previous day can linger until the next day, pressuring employees to cope with their daily job stressors (i.e., job insecurity) and hence reduce the resources that they can invest in their tasks and roles. Third, ethical leader behavior is conceptualized as an important boundary condition that operates to replenish or further exhaust employees' resources in contexts in which frontline service employees experience job insecurity and resource loss during daily work operations. Thus, our study attempts to extend the growing body of literature on the daily behaviors of leaders (Kelemen et al., 2020) .

The growing uncertainty engendered by adverse situations (e.g., the ongoing pandemic and economic downturn) has resulted in the creation of unstable work environments. These changes have implications on the fluctuating level of perceived job insecurity and leader behaviors. Following these notions, in contradistinction to the vast literature on the role of job insecurity as a chronic and continuous job stressor, recent studies in this domain have adopted intraindividual models, which assume that perceived job security levels vary on a daily basis (e.g., Garrido Vásquez et al., 2019; Schreurs et al., 2012) . Specifically, Schreurs et al. (2012) found that, at the daily level, the negative relationship between job insecurity and in-role performance was weaker when employees received supervisor support. In another study, daily coworker conflict was positively related to job insecurity on subsequent days, when between-person job insecurity was controlled for (Garrido Vásquez et al., 2019) . Furthermore, researchers have also recently explored the various forms and consequences of daily leader behaviors based on the notion that leaders interact with followers on a daily basis and are responsible for employee functioning (Keleman et al., 2020) . Indeed, Bormann (2017) found that, at the daily level, the indirect effect of ethical leadership on helping through work engagement was most evident when employees had experienced more abusive supervision the previous day.

Based on the recent implications of the within-person perspective of job insecurity and leader behaviors, we apply COR theory to argue that failure to engender job security will result in greater consumption of the emotional and cognitive resources of employees (Hobfoll et al., 2018) . Thus, an outstretch of employees' resources arising from job insecurity may recur the next day in the form of an intense negative emotion, occupational regret (Zeelenberg et al., 1998) ; this refers to daily negativity generated by comparing current outcomes (i.e., the "what is") and expected outcomes (i.e., the "what could have been" if one had acted differently) of an occupation (Zeelenberg, 1999) . Incorporating a constant decision-making process that involves cognitive appraisals of situations, individuals make perpetual comparisons between factual outcomes and outcomes that could have occurred (Zeelenberg et al., 1998) . As a consequence of daily decision-making, post-decisional negative reactions (i.e., regret) may be observed on a day-to-day basis. This is because the frequency and intensity of a regretted experience are likely to depend on evaluative reactions to daily situations, where employees do not see clear future possibilities to remedy the regret, or no longer perceive future opportunities that were once available (Beike et al., 2009; McMullen & Markman, 2002; Summerville & Roese, 2008) . Following this, employees may strive to preserve the self by reevaluating their current situation and ruminating on limited alternatives, allowing the stressor to pass by reducing work efforts (Hobfoll et al., 2018; Verbruggen & van Emmerik, 2020) . However, regardless of such employees' defensive efforts to preserve the self, perceptions of job stressors and the feeling of regret may linger for a short period of time (e.g., until the next day) (Verbruggen & van Emmerik, 2020) , making it difficult for employees to replenish the resources that were consumed the previous day. Moreover, as intense emotional reactions are elicited, employees may employ coping strategies by reducing the effort that they invest in task activities to conserve resources or explore alternatives from a limited pool of options (Hobfoll et al., 2018) .

Despite the aforementioned negative reaction resulting from perceived job insecurity, employees' daily interactions with leaders, whose behaviors signal high standards of fairness and transparency, can be perceived as both task and emotional resources (Brown & Treviño, 2006; Brown et al., 2005) . Signals provided on a daily basis may render employees less vulnerable to potential resource losses because they are presented with a credible source of resources and leader behavior that prioritizes a follower's best interest in coping with negative reactions to daily job stressors (Lemoine et al., 2019) . Depending on whether employees receive such ethical treatment from leaders, they may either respond favorably by adopting positive work attitudes and being proactive (Lam et al., 2015) or experience more regret and further withdraw resources from task activities. Thus, we propose a moderated mediation model in which the mediated relationship between daily job insecurity, occupational regret, and work outcomes is moderated by daily ethical leader behavior. The proposed research model is presented in Fig. 1 , and each hypothesis is explained in detail in subsequent sections.

In this study, we examine two daily outcomes of job insecurity: work engagement and customer-directed helping. Work engagement refers to high levels of investment in work tasks (Rich et al., 2010) . In contrast, customer-directed helping, which is a primary activity in which frontline employees engage to address customer needs, involves in-role task performance (Chuang & Liao, 2010; Liao & Chuang, 2004) . Past studies have found that daily stressors are related to poor daily work engagement (e.g., Breevaart & Bakker, 2018) and negatively associated with daily organizational citizenship behaviors (e.g., Binnewies et al., 2009) .

We first propose that job insecurity will have negative indirect effects on daily work outcomes through occupational regret the next morning for the following three reasons. This contention is founded upon COR theory (Hobfoll et al., 2018) , in that experiencing negativity (i.e., job insecurity) on a given day will likely render employees with fewer resources and thus make it difficult for them to supplement resources and offset their losses the following day. Following such notions, first, job insecurity on a given day may induce rumination overnight. When employees fail to replenish necessary resources the previous day, they are likely to ponder over the uncertainty caused by intense job stressors and foresee a negative occupational future (Wang et al., 2013) . Because of the lingering effects of the negative emotions elicited by job insecurity, recurrent rumination may trigger thoughts about unattained goals and resources (Baranik et al., 2017) and induce a sense of regret the following day. Second, job insecurity may relate to negative cognitive appraisals of their current circumstances (i.e., subjective assessments of the perceived effect of the context on the self; Lazarus, 1991) . Such evaluations are made when employees believe that their time and effort have been consumed without adequate support to compensate for their losses. This is likely to elicit negative responses to their occupations the following day, and they may result from their poor expectancies of their current contexts (Oreg et al., 2018) . Finally, employees may ponder over forfeited opportunities to acquire resources elsewhere. This is because they perceive that once available resources and opportunities are no longer available and that future options are not apparent or appear limited during daily operations (Beike et al., 2009) . As a result, they may notice a relative lack of opportunities to acquire beneficial resources within their current occupational environment. When employees perceive an unamendable future path that is unlikely to change in the near future, they are likely to experience intense feelings of regret the following workday.

Next, there are several reasons why occupational regret may negatively influence individual outcomes at the daily level. As implied in the defensive principle mentioned above, employees may disengage from work related activities because of two reasons. First, employees are likely to exhibit withdrawal responses to cope with the stress experienced the previous day and to deal with the negative emotions (i.e., regret) experienced before the start of a workday (Ashford et al., 1989) . Employees may decommit from their current organization because of the negativity encountered during daily operations the previous day and thus endorse strategies that help to preserve the limited resources that one might possess after encountering intense negative emotions. When employees decommit, they are less likely to invest daily task effort and may enter into a defensive mode to explore alternative strategies. Consequently, they may be less inclined to invest resources in immediate tasks and may search for opportunities outside their current organization (Hobfoll et al., 2018) . Experiencing extreme negative emotions of regret the following day employees perceive job insecurity, regret may be negatively related to not only their mental resilience but also work engagement because of a lost sense of meaning that was previously derived from their work and occupation (Schaufeli & Bakker, 2004) . A perceived reduction in the meaningful opportunities that are available to employees in their occupation may be detrimental to their job satisfaction and the genuineness of their service interactions with customers (Grandey, 2003) .

Also, intense regret may relate to a heightened level of psychological detachment from daily task activities. Indeed, when employees believe that additional investment of their daily resources will lead to resource depletion and relatively limited gains, they may reevaluate their occupational future and adopt strategies that minimize resource loss during customer interactions. In such situations, they are less likely to engage in daily proactive behaviors that benefit their organizations. Because of the reduction in the available amount of resources to cope with the strain caused by regret, employees may opt to disengage from the process of creating job resources, which may play a critical role in their engagement in daily task activities ( Van den Heuvel et al., 2010) . Without the resources needed to cope with detrimental situations, employees are less likely to go the extra mile to achieve their daily goals. Thus, we hypothesize:

Hypothesis 1 Daily job insecurity has indirect effects on next-day (a) work engagement and (b) customer-directed helping through next-morning occupational regret.

Ethical Leader Behavior refers to actions that influence followers on a daily basis through a principled and moral approach (Lin et al., 2016) , aiming to act as both a "moral person" and a "moral manager" (Brown et al., 2005; Lemoine et al., 2019) . Leaders who engage in ethical behaviors interact with their followers, exhibiting two-way communication, role modeling, care and consideration for follower well-being, and incentivization of normative behaviors (Fehr et al., 2015) . In this sense, leaders promote appropriate work conduct characterized by exemplifying fairness and transparency in interactions with subordinates (i.e., "moral person"), while also serving as credible and legitimate role models to employees ("moral manager"; Fehr et al., 2015) . Thus, employees likely perceive leaders as valuable figures, especially in insecure job situations where resource loss is more salient (i.e., gain paradox principle; Hobfoll et al., 2018) . Daily fluctuations in leader behaviors result in a perceived sense of resource gain or loss and influence access to valuable resources (Zhou et al., 2020) . Accordingly, employee perceptions of ethical leader behavior not only increase employee satisfaction (Brown & Treviño, 2006) , but also foster positive job attitudes and enhance their willingness to invest additional work effort (Bormann, 2017) . In contrast, if employees observe their supervisors engaging in relatively less ethical behaviors and feel deprived of access to meaningful resources, they may regret their past decisions (Beike et al., 2009) . However, because their past choices cannot be rectified immediately through daily work operations, missed opportunities may remain salient to the job-insecure individual. As a result, employees may experience higher levels of occupational regret and are likely to withdraw from task activities.

Ethical leader behavior may influence next-morning occupational regret and next-day job outcomes among jobinsecure employees in two ways. First, from the COR perspective, ethical leaders are a credible source of resource generation and preservation (Brown & Treviño, 2006) . In this sense, despite the possibility of losing their jobs, employees may hold more positive attitudes toward their leaders on workdays (Ng & Feldman, 2015) . Specifically, employees are likely to believe that their leaders will provide them with reliable information on workplace issues (e.g., furloughs or layoffs) and treat them fairly based on accepted rules. Ethical leader behavior replenishes employees' psychological resources, which have been depleted while coping with job insecurity, by strengthening their trust in their leaders (Quade et al., 2019) . Furthermore, ethical leaders provide employees with important resources by giving them clear guidance and alleviating role ambiguity (Brown & Treviño, 2006; Quade et al., 2019) . In turbulent times that involve ethically charged decisions (e.g., layoffs), employees guided by ethical leaders are assured that leaders will not violate ethical standards in their decisions (Quade et al., 2019) ; thus, employees feel less distressed about their future in the organization. Therefore, employees who interact with ethical leaders tend to experience less regret the following morning and reciprocate by investing in task resources (Ng & Feldman, 2015) . However, when employees observe their leaders engaging in less ethical behaviors, they may lower their expectations because of limited access to opportunities and rewards; hence, employees may be treated unfairly (e.g., with regard to decisions on their job status). In such situ, combined with uncertain prospects of their current job, employees may feel more strained and experience greater occupational regret, ultimately devoting less attention to their daily tasks.

Second, exposure to ethical leader behavior may be an important source of social support (i.e., a daily resource) for job-insecure employees (Zhou et al., 2020) . According to Fehr et al. (2015) , a defining characteristic of ethical leader behavior lies in their ability to alleviate suffering and improve well-being among followers. Indeed, when employees observe ethical leader behavior, they are likely to feel secure and confident about their current job, as employees may perceive sufficient emotional and cognitive resources. As a result, employees are more likely to engage in tasks that contribute to service enhancement (Tu & Lu, 2016) . Similar to the effects of exposure to less ethical behavior explained above, a lack of supervisor care will negatively impact the investment of task effort among job-insecure employees. This is because of a heightened sense of uncertainty delivered by their leader behaviors and compromising of follower welfare, thus intensifying their negative reactions toward their jobs. Therefore, when job-insecure employees receive ethical treatment from their leaders, they are likely to believe that they have access to an abundant amount of support. However, when they perceive their leaders to be unreliable sources of social support, they are less likely to hold positive work attitudes that may benefit their organization. Hence, we hypothesize:

Hypothesis 2 Ethical leader behavior moderates the relationship between service employees' daily job insecurity and next-morning occupational regret, such that this relationship is stronger when employees observe less ethical leader behavior the previous day.

Taken together, Hypotheses 1 and 2 indicates a conditional indirect effect: ethical leader behavior moderates the indirect effects of daily job insecurity on next-day work engagement and customer-directed helping through next-morning occupational regret. Hypothesis 1 states that daily job insecurity has a negative effect on daily outcomes through occupational regret the next morning. Hypothesis 2 reflects that the relationship between daily job insecurity and next-morning occupational regret varies as a function of ethical leader behavior. Thus, the indirect effects (Hypothesis 1) are expected to be conditional on ethical leader behavior.

Hypothesis 3 Ethical leader behavior moderates the indirect effects of service employees' daily job insecurity on nextday (a) work engagement and (b) customer-directed helping through next-morning occupational regret, such that these negative indirect effects are stronger when employees observe less ethical leader behavior the previous day.

South Korea reported its first confirmed case of COVID-19 on January 20, 2020. After the number of confirmed cases increased drastically in mid-February, the South Korean government implemented a nationwide social distancing policy on March 22, 2020 (Hwang et al., 2020) . However, lockdown and stay-at-home orders were not issued. Therefore, across most service sectors, frontline employees were required to work during the pandemic. Employees who (a) worked five days a week, (b) had access to the internet, and (c) were working on the frontlines in service organizations were eligible for inclusion and thus to voluntarily participate in surveys. Participants who completed all the surveys received a gift card worth 30 USD for participating. During the first stage of online data collection, demographic and individual characteristics were assessed (i.e., first and second week of March 2020). Two weeks later, they were asked to complete two online surveys a day across a period of five consecutive days (i.e., March 30 to April 3, 2020). Each day, participants were asked to complete a survey before work and one survey after work; the before-work survey had to be completed before starting work, while the after-work survey had to be completed after the end of the workday.

A total of 311 individuals expressed interest in this study and participated in the initial survey. Of these 311 participants, 185 participants provided responses to the before-work and after-work surveys each day. On average, the participants completed the before-work (i.e., morning) survey between 7:22 a.m. and 10:42 p.m. and the afterwork survey between 7:00 p.m. and 11:53 p.m. During each survey period, participants had a two-hour window to complete the survey. Each survey link expired beyond the two-hour window. In accordance with the procedures used in past within-person research studies (e.g., Rosen et al., 2016; Schilpzand et al., 2018) , we followed the suggestion that at least "three data points per person are statistically needed to appropriately model within-person relationships" (Gabriel et al., 2019, p. 92) . Based on this procedure, we excluded participants who did not complete at least three survey days (i.e., participated in one day: n = 34; participated in two days: n = 16). Thus, the final sample included 135 participants who provided 636 observations (94.2% completion rate; 4.7 days per participant) of data of the total possible 675 observations (135 participants × 5 days).

The participants represented diverse sectors within the service industry: retail stores (e.g., department stores and supermarket sales clerks; n = 62, 45.9% of the total sample), banking and financial services (e.g., bank clerks and insurance salespersons; n = 51, 37.8% of the total sample), restaurants (e.g., waiters and kitchen chefs; n = 14, 10.4% of the total sample), and hospitality/tourism (e.g., flight attendants, hotel employees, and tour guides; n = 8, 5.9% of the total sample). With respect to demographic characteristics, 59.3% of the participants were women and their average age was 37.37 years (SD = 8.82). The average organizational tenure was 5.75 years (SD = 4.92), and 86.6% of the participants held at least an undergraduate degree.

We adapted and reworded the measurement items in such a manner that they reflected daily experiences and reactions (Fisher & To, 2012) . All items were rated on a 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree). Further, all survey items were translated into Korean and back-translated into English (Brislin, 1970) .

We measured daily job insecurity at the end of the workday with three items (mean α = 0.98) from De Witte's (2000) and Schreurs et al.'s (2012) job insecurity scale. One sample item is: "Today, I believe that I will lose my job in the near future."

To assess ethical leader behavior, we used six items (mean α = 0.93) from Bormann's (2017) 

We assessed next-morning occupational regret using six items (mean α = 0.85) from Sullivan et al.'s (2007) subjective regret scale. The following is a sample item: "Today, I wish I had gone back to school and restarted another career path that was better suited to me."

To measure daily work engagement, we used Schaufeli et al.'s (2019) three-item (mean α = 0.92) shortened Utrecht Work Engagement Scale (UWES-3). One sample item includes: "Today, at work, I was bursting with energy."

We used Yue et al.'s (2017) three-item (mean α = 0.89) scale to measure daily customer-directed helping. The following is a sample item: "Today, I voluntarily assisted customers even though it meant going beyond my job requirements."

We controlled for baseline (Day 1) positive affect (PA: inspired, attentive, and active; mean α = 0.93) and negative affect (NA: upset, hostile, and nervous; mean α = 0.89), which were assessed using the six-item International Positive Affect and Negative Affect Schedule-Short Form (Thompson, 2007) . Indeed, both PA and NA have been found to be related to regret (Schoemann et al., 2012; Sheehan et al., 2008) , work engagement (Ilies et al., 2017) , and customer-directed helping (Yue et al., 2017) .

Before conducting the main analysis, we computed intraclass correlation (ICC 1 ) coefficients for each variable. They ranged from 55.9% (customer-directed helping) to 79.0% (job insecurity) and justified the adoption of the withinperson approach (see Table 1 ). Therefore, as our data have a multilevel structure (i.e., nested nature of the data), the full hypothesized model was tested by conducting random coefficient modeling with Mplus 8.4 (Muthén & Muthén, 1998 -2017 . Across all models, maximum likelihood (ML) estimation was used to adjust the standard errors of the indicators. To test our hypotheses, we adopted the unconflated multilevel modeling method to avoid biased estimations of indirect effects (Preacher et al., 2010) . The proposed mediation and conditional indirect effects were tested via Monte Carlo simulation procedures using open-source software R (see http:// quant psy. org) to estimate the confidence intervals (CIs) of the indirect effects. We used this method as it yields asymmetric CIs that reflect the skewed nature of the sampling distribution of the indirect effect, thus rendering robust estimations of CIs (Preacher et al., 2010) .

We specified Level-1 (i.e., within-person model) effects of the multilevel path model for the following paths. The effect of daily job insecurity on occupational regret was specified, as well as the effect of occupational regret on next-day work engagement and customer-directed helping, respectively. We also estimated the covariances among the effects to test the indirect effects specified at the within-person model (Bauer et al., 2006) . For the Level-1 moderation term, we specified the fixed effect of job insecurity and one fixed interaction term (i.e., job insecurity and ethical leader behavior). Specifically, the Level-1 interaction terms were created by multiplying the person-mean-centered scores of the exogenous variables (i.e., daily job insecurity and ethical leader behavior) prior to the path analyses. Table 1 presents means, standard deviations, and intercorrelations among the study variables. To evaluate the validity of the measurement model, we conducted multilevel confirmatory factor analysis. The fit indices indicated that the proposed model was a good fit for the data, χ 2 (303) = 594.72, p < 0.05, comparative fit index = 0.94, Tucker-Lewis index = 0.94, root mean square error of approximation = 0.04, and standardized root mean square residual within = 0.06. All factor loadings were significant (range 0.49-0.93).

Hypotheses 1a and 1b proposed that daily job insecurity has indirect effects on next-day work engagement and customer-directed helping through next-morning occupational regret. Controlling for Day 1 PA and NA, daily job insecurity had a significant indirect effect on next-day work engagement through next-morning occupational regret (indirect effect = 0.015, 95% confidence interval [CI] = [-0.035, -0.001]), supporting Hypothesis 1a. Furthermore, the negative relationship between daily job insecurity and next-day customer-directed helping was mediated by next-morning occupational regret, (indirect effect = -0.013, 95% CI [-0.031, -0.001]), supporting Hypothesis 1b. Such results indicate that when individuals experienced more occupational regret the next morning, they were less likely to engage in their work or direct task effort to helping customers during the workday.

Hypothesis 2 posited that the association between daily job insecurity and next-morning occupational regret is stronger among those who report lower levels of ethical leader behavior. As predicted, after controlling for the main effect of daily job insecurity on next-morning occupational regret (γ = 0.10, p < 0.05), the interaction term (i.e., daily job insecurity and ethical leader behavior) was significant (γ = -0.34, p < 0.05; see Table 2 ). We plotted this interaction at high and low levels of (± 1 SD) the moderator. As revealed in Fig. 1 , the link between daily job insecurity and nextmorning occupational regret was stronger among those who reported lower levels of ethical leader behavior (γ = 0.33, p < 0.05). Further, daily job insecurity was not related to next-morning occupational regret among those who reported mean (γ = 0.10, p > 0.05) or high (γ = -0.14, p > 0.05) levels of ethical leader behavior. This indicates that the effect of daily job insecurity on the occupational regret of next morning was greater when employees observed less ethical behaviors of their leaders, thus providing support for Hypothesis 2.

Hypotheses 3a and 3b proposed that daily job insecurity will have a conditional indirect effect on next-day work engagement and customer-directed helping through nextmorning occupational regret. As expected (see Table 3 ), the negative indirect effect of daily job insecurity on nextday work engagement through next-morning occupational regret was significant among those who reported low levels of ethical leader behavior (indirect effect = -0.070, 95% CI [-0.153, -0.011]), but not among those who reported mean (indirect effect = -0.020, 95% CI [-0.092, 0.043]) or high (indirect effect = 0.030, 95% CI [-0.018, 0.093]), levels of ethical leader behavior. These findings supported Hypothesis 3a. Furthermore, the negative indirect effect of daily job insecurity on next-day customerdirected helping through next-morning occupational regret was significant among those who reported low levels of ethical leader behavior (indirect effect = -0.059, 95% CI [-0.113, -0.005]), but not among those who reported Table 1 Descriptive statistics and intercorrelations among the study variables ICC 1 = interclass correlation coefficient (1). The mean values of Cronbach's alpha (across days) are presented along a diagonal. The correlations presented above the diagonal represent within-person correlations (N = 501). The correlations presented below the diagonal represent betweenperson correlations (N = 135). Composite reliability is computed as the squared sum of the factor loadings divided by the sum of the squared sum of the factor loadings and sum of the error variance * p < 0.05 (two-tailed), ** p < 0.01 (two-tailed) These results indicated that daily job insecurity was negatively and significantly related to the work engagement and customer-directed helping the next day via occupational regret experienced in the morning of the next day. The results are summarized in Fig. 2 .

The present findings, which suggest that job insecurity is a prevalent issue that employees face on a daily basis during adverse situations (e.g., COVID-19 pandemic and economic downturns), have crucial implications for organizations and leaders. In accordance with the COR theory, daily job insecurity had negative indirect effects on two outcomes that are critical to service organizations: next-day work engagement and customer-directed helping. Further, daily job insecurity negatively influenced these daily outcomes through (increased) occupational regret the next morning. These detrimental effects were more pronounced among individuals who reported lower levels of ethical leader behavior. This study makes several contributions to the literature. First, emergent within-person relationships are consistent with the findings of recent studies examining the detrimental effects of job insecurity on employees at the daily level (e.g., Breevaart & Tims, 2019; Garrido Vásquez et al., 2019; Schreurs et al., 2012) . This perspective emphasizes that work contexts have become increasingly volatile (e.g., organizational restructuring, external competition, and flexibility demands), which may have unintended negative consequences for employees. Despite a general agreement on the negative impact of job insecurity at the betweenperson level (Sverke et al., 2002) , investigating daily variations in job insecurity is important for a better understanding of employees' short-term emotions and behaviors, as employees are prone to fluctuating work demands and social interactions (Fisher & To, 2012) . This study specifically focused on the emotional and behavioral responses of frontline employees who worked regularly during the ongoing COVID-19 pandemic. We suggest that employee reactions are likely a function of a highly volatile context, underscoring that the intraindividual aspect of perceived job insecurity can assist our understanding of withinperson fluctuations in employees' situations. In contrast to a between-person entity perspective that assumes job insecurity to be a chronic job stressor and that individuals form a certain pattern of job perception, our findings imply that stressors are crucial in the daily functioning of individuals and organizations. Hence, by adopting a within-person perspective, we were able to demonstrate that frontline employees who work under uncertain conditions disengage from their jobs to preserve their internal resources (Hobfoll et al., 2018) . Relatedly, to observe employees placed in extreme environments that entail uncertainty, this study was conducted in South Korea after the outbreak of COVID-19. This study provides a relatively stringent test of employees' daily experiences in an adverse context (e.g., COVID-19, economic turmoil, or natural disasters), as frontline employees faced high levels of volatility. Such a test corresponds with studies on employee adversity and traumatic events, where a specific site was selected for research (e.g., the aftermath of the September 11, 2001 attacks or fire victims; Dutton et al., 2002 Dutton et al., , 2006 . Past studies targeting such traumatic events have repeatedly shown that such events take a toll on workplaces by requiring employees to invest additional time and effort to deal with emerging demands and uncertainties. The COVID-19 pandemic highlights a novel context, especially for frontline service employees. Involving imminent threats for frontline employees facing changing shutdown and social distancing policies, in addition to furloughs and layoffs in workplaces, such outbreaks are specific to infectious diseases and thus present additional struggles for employees who interact face-to-face with customers. Considering that the COVID-19 pandemic has had tragic circumstances on frontline employees and has disrupted their daily routines, it is crucial to unravel the day-to-day evaluations, emotions, and behaviors of the employees during these events. Hence, our research in the work context during the early periods of COVID-19 allowed us to determine the time-lagged effects of employees' job stressors on their next-day work behaviors.

Second, we examined the mediating role of nextmorning occupational regret, a specific type of emotional reaction that partially explains why job insecurity is negatively associated with daily work outcomes. Frontline service employees have been regarded as one of the occupational groups to have been hit hardest since the outbreak of COVID-19, confronting constant layoffs and furloughs Oh & Lee, 2020) . As researchers have claimed that job insecurity is indicative of "the subjective anticipation of an involuntary event" (Jiang & Lavaysse, 2018 , p. 2308 such as COVID-19, it is imperative to consider both the perceived threat to one's job, and emotional reactions to job insecurity (Sverke et al., 2002) . Research has suggested that emotional reactions to perceived job insecurity (i.e., negative evaluation of one's job) are more closely related to employee outcomes, implying that a negative assessment in response to job insecurity may serve as a mechanism between job insecurity and employee outcomes (Jiang & Lavaysse, 2018) . Our findings address these points by indicating that job stressors that create a sense of uncertainty have lagged effects on daily work outcomes through strong reactions of occupational regret that extend to the following day. In particular, the prolonged negative effects of job insecurity on daily work outcomes was strengthened by increased next-morning occupational regret. Thus, the findings of this study enhance our understanding of employees who exhibit negative appraisals and emotions toward their job, as they are unable to remain committed to their current occupation because of previous-day job stressors and, consequently, withdraw their contributions to organizational operations (Costa & Neves, 2017) . Finally, by examining ethical leader behavior as a critical boundary condition that impacts the effects of perceived job insecurity, this study extends the literature on moral and ethical factors within the workplace (e.g., Dasborough et al., 2020; Lemoine et al., 2019) . Past studies have focused on how and when leaders can play a supportive role in stressful situations by exhibiting servant or authentic leadership (Schreurs et al., 2012; Wu et al., 2020) , thereby equipping followers with the necessary task and social resources. We go a step further to propose and test that not only is presenting social resources in day-to-day situations important but also leaders' engagement in normative behaviors helps frontline employees cope with threats to their job security (Bormann, 2017; Lin et al., 2016) . In other words, we go beyond simply viewing leaders as figures who provide employees with task and social resources. Instead, our study also assesses whether it is important for employees to perceive that their leaders comply with normative standards, maintain fairness, and engage in appropriate behavior (Wang et al., 2021) . Our findings suggest that ethical leader behavior functions as a key resource that helps employees cope with job insecurity perceptions and protects them from having skeptical thoughts about their job, while a lack of ethical leader behavior may elicit followers' perception of a deficiency in work resources, rendering them vulnerable to job security threats. Thus, the present findings are distinct from prior research on the role of leader behavior (Kelemen et al., 2020) in the context of job insecurity, by demonstrating ethical leader behavior as a crucial form of leader action acts as a buffer for employees against the negative consequences of job insecurity perceptions.

The present findings offer several suggestions that leaders and organizations can use to effectively combat the detrimental effects of adverse contexts and the perceptions of employees' uncertain job status on work outcomes. Notably, when employees are not adequately exposed to ethical leader behavior on a given day, they may find it difficult to cope with the adverse effects of daily job insecurity. This may intensify their feelings of occupational regret and cause them to further disengage from their daily work tasks the next day. Therefore, leaders should engage in ethical behaviors toward frontline employees. This will inform employees that the psychological contract between their supervisors and them is intact and that it would be worthwhile for them to continue engaging in work activities (Brown & Treviño, 2006) . Considering that employees inevitably experience fluctuating levels of job insecurity during crisis situations, it may be imperative for leaders to create an environment that upholds established ethical principles through compassionate interactions and fair treatment of their employees. The present findings reinforce the notion that, even though stressors such as job insecurity may unavoidably have deleterious effects on frontline employees during an ongoing pandemic, supervisors can effectively manage such employees by providing them with adequate resources (e.g., task and emotional support; Lim, 1996; Schreurs et al., 2012) .

Based on the finding that perceived job insecurity undermines employees' work engagement and customer-directed helping behavior through occupational regret, it is critical to reduce perceived job insecurity. Job insecurity scholars have highlighted that perceptions of job insecurity can be alleviated by clarifying performance expectations and criteria (Schreurs et al., 2012; Shin & Hur, 2021) . In addition, organizations can reduce perceived job insecurity among employees by boosting their motivation and self-efficacy through goal setting, job design, and training (Lam et al., 2015; Shin & Hur, 2021) , which are also helpful in decreasing occupational regret. Furthermore, given that occupational regret directly exerts an adverse effect on customer-directed helping, organizations and leaders must recognize regret as a strong emotion that may impede employee functioning related to customer service. Indeed, the extent to which frontline employees deliver high-quality services appears to determine customer satisfaction and service performance (Bowen et al., 1989) . Influenced by job insecurity and amplified by feeling occupational regret, it may be difficult for employees to concentrate on their core service behaviors during the pandemic. Although this may be applicable to employees across different sectors, frontline employees may be among the worst affected groups. Thus, organizations should create interventions that reduce frontline employees' occupational regret. This can be realized by building a work environment that provides employees with social support (e.g., caring and compassion) and alleviates job stressors (Dutton et al., 2006; Huang et al., 2017) .

This study has some limitations, which can be addressed in future studies. First, although self-report measures are commonly used in daily research studies, they are vulnerable to common method bias (Podsakoff et al., 2012) . Our adoption of a repeated measures design based on the experience sampling method is likely to have reduced potential recall biases. Further, we aimed to temporally separate the measurement of all study variables to address these concerns. Moreover, the results of confirmatory factor analysis supported the construct validity of all the measures used in this study. However, in future studies, data could be collected from multiple sources (e.g., supervisor and coworker evaluations of employee performance behaviors).

The second limitation pertains to the limited timeframe within which the surveys were conducted (i.e., after the first outbreak of COVID-19 and before the implementation of the first nationwide social distancing policy in South Korea). Although this study was planned to observe employees in a highly distressed environment, it also captured a certain period during the upstart of the pandemic. In this sense, the results found in this study may be stronger or weaker based on the COVID-19 severity context (e.g., COVID-19 cases and deaths, economic hardships, and national lockdown policies). Similarly, the study design did not permit us to compare the effects of job insecurity on work outcomes before and after the outbreak. Nonetheless, based on an observation of employees spanning five consecutive workdays, we were able to determine their daily experience of job stressors and emotional and behavioral reactions.

Third, another limitation concerns the examination of boundary conditions that may contribute to reversing the negative effects of perceived job uncertainty on prosocial behaviors. In contrast to the study findings, we acknowledge that frontline employees may increase their engagement in customer-directed helping or coworker-directed helping behavior, particularly in situ. Thus, there may be conditions under which perceived job insecurity is positively related to prosocial behavior. A promising starting point for investigating these conditions could be how the job is designed (Hackman & Oldham, 1975) . For example, the extent to which frontline employees perceive their tasks as meaningful and significant to their organization's functioning may help alleviate or even reverse the negativity of job insecurity. Additionally, when presented with adequate job autonomy, an individual may be determined to change their job boundaries to enhance work conditions and thus reinforce prosocial behaviors.

Fourth, although we identified ethical leader behavior as a critical form of leader behavior that protects employees from the negative consequences of job insecurity, there may be other leadership variables (e.g., servant or authentic leadership) that potentially moderate the job insecurity-work outcome relationship. As we did not analyze the relative importance of diverse forms of leader behavior, we cannot ascertain whether ethical leader behavior is more important in job-insecure situations than other forms of leader behavior. Thus, for a more elaborate understanding of leader behavior in the domain of job insecurity, it is necessary to test the effect of ethical leader behavior after controlling for other forms of leader behavior. Relatedly, how ethical leader behavior mitigates or strengthens the deleterious effect of job insecurity needs to be further elucidated in future research. It would be intriguing to explore the specific mechanisms (e.g., perceived fairness, trust, and leader-member exchange) through which ethical leader behavior acts as a buffer against job insecurity.

Finally, we examined only one daily mechanism that explains the indirect effects of daily job insecurity on work outcomes during a crisis. Although occupational regret is an important negative response, the significant direct path that extended from daily job insecurity to next-day work engagement warrants further exploration of additional mechanisms.

For example, future studies should examine the effect of resilience, which confers both emotional and cognitive resources, on the emotional toll that job insecurity takes on individuals (Shoss et al., 2018) . In addition, daily behavioral mechanisms such as job crafting may predict employee effort expenditure in response to negative changes in their work context (Boswell et al., 2014) .

Content, causes, and consequences of job insecurity: A theory-based measure and substantive test

Customer mistreatment, employee health, and job performance: Cognitive rumination and social sharing as mediating mechanisms

Conceptualizing and testing random indirect effects and moderated mediation in multilevel models: New procedures and recommendations

What we regret most are lost opportunities: A theory of regret intensity

Daily performance at work: Feeling recovered in the morning as a predictor of day-level job performance

Linking daily ethical leadership to followers' daily behaviour: The roles of daily work engagement and previous abusive supervision

I cannot afford to have a life: Employee adaptation to feelings of job insecurity

A framework for analyzing customer service orientations in manufacturing

Daily transactional and transformational leadership and daily employee engagement

Daily job demands and employee work engagement: The role of daily transformational leadership behavior

Crafting social resources on days when you are emotionally exhausted: The role of job insecurity

Back-translation for cross-cultural research

Ethical leadership: A review and future directions

Ethical leadership: A social learning perspective for construct development and testing

Strategic human resource management in service context: Taking care of business by taking care of employees and customers

Job insecurity and work outcomes: The role of psychological contract breach and positive psychological capital

The generation and function of moral emotions in teams: An integrative review

Arbeidsethos en jobonzekerheid: Meting en gevolgen voor welzijn, tevredenheid en inzet op het werk [Work ethic and job insecurity: Measurement and consequences for well-being, satisfaction, and performance

Leading in times of trauma

Explaining compassion organizing

Moralized leadership: The construction and consequences of ethical leader perceptions

Using experience sampling methodology in organizational behavior

Do co-worker conflicts enhance daily worries about job insecurity: A diary study

Experience sampling methods: A discussion of critical trends and considerations for scholarly advancement

Why essential workers deserve Covid-19 hazard pay

When "the show must go on": Surface acting and deep acting as determinants of emotional exhaustion and peerrated service delivery

Development of the Job Diagnostic Survey

Conservation of resources in the organizational context: The reality of resources and their consequences

Deviance and exit: The organizational costs of job insecurity and moral disengagement

Emotional exhaustion among the South Korean workforce before and after COVID-19. Theory, Research and Practice

Why do employees have better family lives when they are highly engaged at work?

Cognitive and affective job insecurity: A meta-analysis and a primary study

Leading day-to-day: A review of the daily causes and consequences of leadership behaviors

The performance implications of job insecurity: The sequential mediating effect of job stress and occupational commitment, and the buffering role of ethical leadership

Job insecurity and subjective sleep quality: The role of spillover and gender. Stress and Health. Advance online publication

The effect of the COVID-19 pandemic on South Korean domestic industry. Korean Institute for Industrial Economics & Trade

Job insecurity and organizational citizenship behavior: Exploring curvilinear and moderated relationships

Emotion and adaptation

COVID-19 Business Report. Samjong KMPG Economic Research Institute

Taking stock of moral approaches to leadership: An integrative review of ethical, authentic, and servant leadership

The influence of social comparison on career decision-making: Vocational identity as a moderator and regret as a mediator

A multilevel investigation of factors influencing employee service performance and customer outcomes

Job insecurity and its outcomes: Moderating effects of work-based and nonwork-based social support

When ethical leader behavior breaks bad: How ethical leader behavior can turn abusive via ego depletion and moral licensing

Coping with job insecurity: The role of procedural justice, ethical leadership and power distance orientation

Affective impact of close counterfactuals: Implications of possible futures for possible pasts

Mplus user's guide

Ethical leadership: Metaanalytic evidence of criterion-related and incremental validity

The measure and assessment of the employment vulnerability during COVID-19 (BOK Issue Note No

An affect-based model of recipients' responses to organizational change events

Sources of method bias in social science research and recommendations on how to control it

A general multilevel SEM framework for assessing multilevel mediation

Boundary conditions of ethical leadership: Exploring supervisor-induced and job hindrance stress as potential inhibitors

Job engagement: Antecedents and effects on job performance

Who strikes back? A daily investigation of when and why incivility begets incivility

Job demands, job resources, and their relationship with burnout and engagement: A multi-sample study

An ultra-short measure for work engagement: The UWES-3 validation across five countries

Not too tired to be proactive: Daily empowering leadership spurs next-morning employee proactivity as moderated by nightly sleep quality

Regrets, I've had a few: Effects of dispositional and manipulated attachment on regret

A weekly diary study on the buffering role of social support in the relationship between job insecurity and employee performance

The ups and downs of felt job insecurity and job performance: The moderating role of informational justice

Regret associated with the decision for breast reconstruction: The association of negative body image, distress and surgery characteristics with decision regret

When do job-insecure employees keep performing well? The buffering roles of help and prosocial motivation in the relationship between job insecurity, work engagement, and job performance

Bending without breaking: A two-study examination of employee resilience in the face of job insecurity

No regrets? An investigation of the relationship between being laid off and experiencing career regrets

Dare to compare: Fact-based versus simulation-based comparison in daily life

No security: A metaanalysis and review of job insecurity and its consequences

Development and validation of an internationally reliable short-form of The Positive and Negative Affect Schedule (PANAS)

Do ethical leaders give followers the confidence to go the extra mile? The moderating role of intrinsic motivation

Personal resources and work engagement in the face of change

The effect of job insecurity on employee health complaints: A within-person analysis of the explanatory role of threats to the manifest and latent benefits of work

When staying is dissatisfying: Examining when and why turnover cognitions affect stayers' career satisfaction

Can't get it out of my mind: Employee rumination after customer mistreatment and negative mood in the next morning

Not all followers socially learn from ethical leaders: The roles of followers' moral identity and leader identification in the ethical leadership process

The relationship between challenge and hindrance stressors and emotional exhaustion: The moderating role of perceived servant leadership

Feeling bad and doing good: The effect of customer mistreatment on service employee's daily display of helping behaviors

Anticipated regret, expected feedback and behavioral decision making

Reconsidering the relation between regret and responsibility. Organizational Behavior and Human Decision Processes

Ethical leadership as the reliever of frontline service employees' emotional exhaustion: A moderated mediation model