Hospitality Industry Employees’ Intention to Stay in Their Job after the COVID-19 Pandemic

: The COVID-19 pandemic had a devastating effect on the tourism and hospitality industries in Taiwan, causing some small companies to cease trading and large companies to place their employees on unpaid leave. Placing employees on unpaid leave may have negatively affected the intention of hospitality employees to remain in their jobs. This study examined whether employees’ job insecurity and organizational identiﬁcation affected their intention to stay in their job during the COVID-19 pandemic. Previously developed scales were adopted to develop items measuring job insecurity, organizational identiﬁcation, and intention to stay in a job. Responses to 515 returned questionnaires were examined. The results revealed that job insecurity signiﬁcantly affects organizational identiﬁcation. Both job insecurity and organizational identiﬁcation signiﬁcantly affected intention to stay. Few studies have used path analyses to investigate the relationships among intention to stay, job insecurity, and organizational identiﬁcation. The indirect effect of organizational identiﬁcation was analyzed, and evidence supporting a total effect and total indirect effect was obtained. This implies that hospitality companies seeking to retain staff during crises should promote organizational identiﬁcation among staff.


Introduction
The outbreak of COVID-19 in 2020 created challenges and difficulties worldwide, compromising numerous industries. Among said industries, the tourism, aviation, and hospitality industries were the first to bear the brunt. To mitigate the effects of the pandemic, Taiwan began implementing border control, which prohibited foreign tourists from entering Taiwan. Such a measure ultimately devastated Taiwan's tourism and hospitality industries. Across Taiwan, numbers of domestic and foreign tourists decreased sharply because of the pandemic, and determining how to survive the challenges of such a period is a critical matter for all stakeholders in the hospitality industry, particularly employees. To address this concern, hospitality businesses have employed a series of approaches such as closing some guest rooms, initiating numerous promotions, negotiating rent reductions, and launching food and beverage delivery services.
According to the Tourism Statistics Database of the Taiwan Tourism Bureau (2021), the number of people visiting Taiwan was 694,187 in 2020, which was a 90% drop (i.e., decrease by 7,749,837) compared with that in 2019 (i.e., 8,444,024). This drastic decrease in the number of tourists illustrates the strong negative effects on Taiwan's hospitality industry, in which many small businesses were forced to suspend their operations or even close down completely. By contrast, large businesses were compelled to streamline their personnel, mandating that their employees had to take unpaid leave. These contingency plans generated a sense of job insecurity (hereafter abbreviated as "job insecurity") in employees, changing their intention to remain and sense of trust in their company. Thus, identification of the intention of hospitality employees to stay in their job and their sense of trust in their company during such a crisis period merits investigation.
However, studies have suggested that the relationships among employees' intention to stay in their job, job insecurity, and organizational identification are ambiguous. Furthermore, most relevant studies have not focused on the unique context of the COVID-19 pandemic era. Therefore, this study fills the gap in the literature on strategies for the retention of hospitality employees during the COVID-19 pandemic. Blau (1964) proposed social exchange theory, which holds that an organization member's relationship with the organization is based on a mutual benefit principle, and relevant rights and obligations are developed through dialog between the member and the organization. Social exchange theory is commonly used to explain the relationship between employees and supervisors or employees and organizations. Based on social exchange theory, job insecurity is a crucial factor influencing the stability of businesses and organizations. Chen and Eyoun (2021) indicated that the increasing numbers of food and beverage industry employees being fired or forced to take leave during the COVID-19 pandemic have elevated employees' fears of losing their jobs and their sense of uncertainty for the future, leading to greater job insecurity. The COVID-19 catastrophe, coupled with the uncertain future of the tourism industry and the decline in tourism demand, has caused hotel employees to experience job insecurity. Furthermore, downsizing by organizations instills the fear of unemployment in employees, increasing their job insecurity Jung et al. (2021). Therefore, in the hospitality industry, personnel and cost-related changes following the pandemic-such as requesting that employees take or accept unpaid leave, job transfers, and reduced salaries-have caused widespread job insecurity among hospitality industry employees.
When organizations are faced with drastic and environment-changing events, organizational identification plays a critical role in bringing organizational members together. Strengthening employees' identification with their organization can help the organization remain competitive and develop sustainably (Chen and Tseng 2020). Thus, enhancing employees' identification with their organization to boost their cohesiveness during the COVID-19 pandemic is critical to ensuring that the employees work with their organization to overcome any difficulties encountered.
Employees' intention to stay in their job indicates whether they still wish to work for their organization after giving the organization their energy and effort. Factors influencing employees' intention to stay include personal factors, work-related factors, and workingenvironment-related factors (Naim and Lenkla 2016). Hsiao and Lai (2015) asserted that providing customer service is the top priority for the tourism and hospitality industries. Because employees in such industries must respond to various customer demands, they experience more intense labor and stronger emotions than do employees in other industries. Thus, enhancing such employees' identification with the operations of their organization can ensure that they work diligently for their organization, ultimately strengthening their intention to stay. To retain more employees, managers can improve systems and methods in their organization by making relevant adjustments so that employees identify with and continue to serve their organization (Lee et al. 2018).
The hospitality industry is a labor-intensive industry with many employment opportunities. Because people who work in this industry can switch company easily, the industry has high employee turnover rates, making personnel shortages a common phenomenon. Thus, determining whether employees' intention to stay in their job during the COVID-19 pandemic differs from that otherwise is worthy of investigation. Accordingly, this study examined whether employees' job insecurity and organizational identification affected their intention to stay in their job during the COVID-19 pandemic.
In accordance with the aforementioned research motivation, this study attempted to answer the following questions:

1.
Does hospitality industry employees' job insecurity have a significant effect on their intention to stay in their job during the COVID-19 pandemic? 2.
Does hospitality industry employees' job insecurity have a significant effect on their organizational identification during the COVID-19 pandemic?
3. Does hospitality industry employees' organizational identification have a significant effect on their intention to stay in their job during the COVID-19 pandemic? 4.
Does job insecurity affect intention to stay through organizational identification?
The answers to these questions would serve as a reference for hospitality companies wishing to retain their employees.

Job Insecurity
Job insecurity is defined as the threat to their current job perceived by an employee (Erdogan et al. 2020). Some scholars have also defined job insecurity as an employee's willingness to continue working or their views on being unemployed during a crisis and an employee's continual fear of losing their job. Job insecurity not only includes employees' fear of unemployment but also their anxiety about threats that may compromise their job status (Ali et al. 2021). Accordingly, this study defined job insecurity as the sense of insecurity that hospitality industry employees feel about their job and the anxiety they experienced about losing their job during the pandemic.

Organizational Identification
Organizational identification occurs when employees consider themselves as members of their organization and define themselves as a part of the group (Teresi et al. 2019). When employees perceive their values to be similar to those of their organization, they identify with their organization (Bharadwaj and Yameen 2020). When the line between employees and their organization becomes blurred, the employees have become part of their organization and agree with its values, goals, and norms (Lee et al. 2015). Therefore, this study defined organizational identification as the degree to which the values and goals of a hospitality industry employee are identical to those of their organization.

Intention to Stay
When an organization gives employees the opportunity to develop themselves and enhance their professional abilities, the employees establish emotional connections with, and become committed to, their organization. These feelings subsequently become the employees' intention to stay in their job (Naim and Lenkla 2016). The emotional connection between an employee and their organization, or even an employee identifying with their organizations, directly and markedly affects the employee's intention to stay stay (Valéau et al. 2019). Therefore, this study defined intention to stay as the emotional connection between an employee and their organization or the employee identifying with their organization, ultimately leading to an intention to stay.

Job Insecurity, Organizational Identification, and Intention to Stay-Related Studies
Few studies have used path analyses to investigate the relationship among intention to stay, job insecurity, and organizational identification. In this study, we also provided a conceptual model to explain why job insecurity exerts a significant positive effect on employees' organizational identification and intention to stay in their present jobs during the COVID-19 pandemic.

Job Insecurity and Organizational Identification
Recent studies on job insecurity have supported the social exchange theory, showing that in general, employees' job insecurity reduces their job engagement and organizational identification, suggesting that job insecurity is negatively correlated with their job engagement and organizational identification (Ngo et al. 2013;Callea et al. 2016;Chirumbolo et al. 2017). Studies have also indicated that job insecurity has a negative effect on organizational identification, resulting in decreased task performance and lower performance levels. Piccoli et al. (2017) surveyed 201 employees and confirmed the significant negative effect of job insecurity on organizational identification. Callea et al. (2016) conducted a study on the effects of job insecurity on organizational identification and job performance and discovered that job insecurity has a significant effect on organizational identification. Accordingly, this study posited H1: Employees' job insecurity would have a significant effect on their intention to stay in their job.

Organizational Identification and Intention to Stay
The social exchange theory posits that employees' strong organizational identification is positively correlated with their intention to stay because organizational identification is employees' views of the common destiny they share with their organizations. Rolf et al. (2004) stated that group members' identities in their respective groups are also their crucial social identities. Bharadwaj and Yameen (2020) noted that organizational identification is a critical factor influencing employees' intention to stay. Bharadwaj and Yameen (2020) surveyed business employees and found that organizational identification had a significant positive effect on intention to stay. The same finding was obtained by Kim et al. (2020) through a survey of 203 business employees. Accordingly, this study posited H2: Employees' job insecurity would have a significant effect on their organizational identification.

Job Insecurity and Intention to Stay
Regarding the effects of employees' job insecurity on their turnover intention, most scholars believed that employees' job insecurity increases their turnover intention and decreases their intention to stay. Ashford et al. (1989) asserted that employees' job insecurity increases their turnover intention. Sverke et al. (2002) remarked that employees' job insecurity has a significant effect on enhancing their turnover intention. Hellgren et al. (1999) conducted a longitudinal study on 375 employees in a Swedish company by using a twodimensional, both qualitative and quantitative, job insecurity scale, where the company was in the process of downsizing. The results showed that after controlling positive and negative affective factors, job insecurity was positively correlated with turnover intention. Ribeiro et al. (2016) discovered that job insecurity increases intention to leave. Thus, job insecurity can be concluded to significantly and negatively affect intention to stay. Fan and Wang (2013) studied bank employees and indicated that the higher a person's job security, the weaker their intention to leave their job. Thus, the higher is job insecurity, the stronger the intention to leave and the weaker the intention to stay. Zhao and Liang (2015) discovered that new employees' job insecurity had a significant negative effect on their intention to stay. Accordingly, this study posited H3: Employees' organizational identification would have a significant effect on their intention to stay in their job.
Because the aforementioned studies noted the significant effect of job insecurity on organizational identification and that of organizational identification on intention to stay, this study investigated whether organizational identification mediated the effect of job insecurity on intention to stay. Accordingly, this study posited H4: Employees' job insecurity would affect their intention to stay in their job through the mediation of organizational identification.
With regard to the above hypotheses, a conceptual model was developed to illustrate the connections among job insecurity, organizational identification, and intention to stay, as displayed in Figure 1.

Sampling Method and Questionnaire Design
This study used structured questionnaires scored using a 7-point Likert scale, where 1 and 7 represented strongly disagree and strongly agree, respectively. Questionnaire items were created after referencing scales with favorable reliability and validity in related studies. The questionnaire was pretested before analysis by using the expert review method. The job insecurity items were adapted from the scale introduced by Ali et al. (2021). The items measuring organizational identification were revised from the scale proposed by Lyngdoh et al. (2018). The intention to stay items were adapted from the scale presented by Tsai and Chang (2015). All three dimensions contained seven items, resulting in a total of 21 items. The second part of the questionnaire collected the hospitality industry employees' demographic information-sex, age, marital status, educational level, job seniority (their number of years in their job), job level, monthly income, and job content.
The questionnaire was distributed using an online Google form to employees of starrated hotels throughout Taiwan from March through June, 2021. A total of 550 questionnaires were issued, of which 530 were returned and 515 were valid. We evaluated for possible common method bias by conducting Harman's one-factor test (Podsakoff and Organ 1986), and the results suggested no common method bias problem was present. On the basis of the survey data, multivariate analyses including frequency distribution, descriptive statistics, and regression were conducted. Bootstrapping was employed to analyze the mediating effects of variables. Cronbach's α was calculated to assess internal consistency.

Analysis Method
This study used SPSS 24 in analysis. Multivariate was conducted, including frequency distribution, descriptive statistics, and path analyses. Subsequently, the Bootstrap method was employed to analyze the mediating effects of variables. We assessed the normality, independence, and constant variance (Schmidt and Finan 2018) of the error term in regression analysis. When work insecurity was considered the independent variable and organizational identity the dependent variable, the Durbin-Watson (DW), Kolmogorov-Smirnov (KS), and Shapiro-Wilk (SW) test results were 1.912, 0.090, and 0.113, respectively. When work insecurity and organizational identity were considered independent variables and intention to stay was a dependent variable, the DW, KS, and SW test results were 2.065, 0.235, and 0.437, respectively. DW was close to 2, suggesting the independence of the model specification, and the KS and SW values indicated that normality assumptions were

Sampling Method and Questionnaire Design
This study used structured questionnaires scored using a 7-point Likert scale, where 1 and 7 represented strongly disagree and strongly agree, respectively. Questionnaire items were created after referencing scales with favorable reliability and validity in related studies. The questionnaire was pretested before analysis by using the expert review method. The job insecurity items were adapted from the scale introduced by Ali et al. (2021). The items measuring organizational identification were revised from the scale proposed by Lyngdoh et al. (2018). The intention to stay items were adapted from the scale presented by Tsai and Chang (2015). All three dimensions contained seven items, resulting in a total of 21 items. The second part of the questionnaire collected the hospitality industry employees' demographic information-sex, age, marital status, educational level, job seniority (their number of years in their job), job level, monthly income, and job content.
The questionnaire was distributed using an online Google form to employees of star-rated hotels throughout Taiwan from March through June, 2021. A total of 550 questionnaires were issued, of which 530 were returned and 515 were valid. We evaluated for possible common method bias by conducting Harman's one-factor test (Podsakoff and Organ 1986), and the results suggested no common method bias problem was present. On the basis of the survey data, multivariate analyses including frequency distribution, descriptive statistics, and regression were conducted. Bootstrapping was employed to analyze the mediating effects of variables. Cronbach's α was calculated to assess internal consistency.

Analysis Method
This study used SPSS 24 in analysis. Multivariate was conducted, including frequency distribution, descriptive statistics, and path analyses. Subsequently, the Bootstrap method was employed to analyze the mediating effects of variables. We assessed the normality, independence, and constant variance (Schmidt and Finan 2018) of the error term in regression analysis. When work insecurity was considered the independent variable and organizational identity the dependent variable, the Durbin-Watson (DW), Kolmogorov-Smirnov (KS), and Shapiro-Wilk (SW) test results were 1.912, 0.090, and 0.113, respectively. When work insecurity and organizational identity were considered independent variables and intention to stay was a dependent variable, the DW, KS, and SW test results were 2.065, 0.235, and 0.437, respectively. DW was close to 2, suggesting the independence of the model specification, and the KS and SW values indicated that normality assumptions were not rejected. The standardized residuals and spread of predicted values revealed no evidence of heteroscedasticity. Therefore, the data used in this study exhibited acceptable independence, normality, and homoscedasticity, and regression analysis was justified.

Sample Data
A total of 515 returned questionnaires were examined in this study. Regarding sex, 321 respondents were female (62.33%). Those aged 31-40 years accounted for most of the sample (258, 50.1%). Unmarried respondents were predominant (278, 53.98%). The vast majority of respondents were a university graduate (388, 75.34%). Regarding seniority, the most common number of years in the job was 3-5 years (188, 36.5%). More than half were basic-level employees (337, 65.44%). The most common monthly income was NT$25,001-50,000 (326, 63.3%). Regarding job content, most of the respondents did work that was not directly related to sales (313, 60.78%).

Descriptive Statistics
According to Table 1, the means and standard deviations of the item scores ranged from 4.24 to 5.03 and from 1.17 to 1.53, respectively. For job insecurity, the internal consistency (Cronbach's α) of the seven items was 0.93. For organizational identification and intention to stay, Cronbach's α was 0.94 and 0.94, respectively.

Path Analyses
As shown in Table 2, t value of job insecurity's effect on organizational identification was 9.453 (p = 0.000 < 0.05), indicating that job insecurity had a significant effect on organizational identification, supporting H1. Job insecurity explained 14.8% of the variance in organizational identification. The t value of job insecurity's effect on intention to stay was 3.525 (p = 0.000 < 0.05), showing that job insecurity had a significant effect on intention to stay, supporting H3. The t value of organizational identification's effect on intention to stay was 14.945 (p = 0.000 < 0.05), revealing that organizational identification had a significant effect on intention to stay, supporting H2. Job insecurity and organizational identification explained 38.6% of the variance in intention to stay.

Intermediary Effect Analysis
Organizational identification mediated the effect of job insecurity (independent variable) on intention to stay (dependent variable), making organizational identification a mediating variable in this study. Indirect effect and confidence interval tests were performed to verify whether the mediating effect was present.
According to the analyses of the indirect effects of the mediating variable, as shown in Table 3, the total effect of job insecurity on retention intention had p < 0.05 and a confidence interval that did not include 0 (i.e., 0.234-0.423), supporting the existence of the total effect. The total indirect effect of job insecurity on retention intention through organizational identification had p < 0.05 and a confidence interval that did not include 0 (i.e., 0.144-0.274), supporting the existence of the total indirect effect and validating the mediating effect of organizational identification on the relationship between job insecurity and intention to stay.

Conclusions
In this study, we examined whether employees' job insecurity and organizational identification affected their intention to stay in their present job during the COVID-19 pandemic. We compared our results with those reported in related literature. Our results revealed that job insecurity had a significant positive effect on organizational identification during the pandemic, a finding that contradicts those of Piccoli et al. (2017) and Callea et al. (2016). We conjecture that during the pandemic, job insecurity was presumably influenced by the government's policies, and thus employer and employee cohesiveness was strengthened, leading to the positive causal relationship between job insecurity and organizational identification; similar to the findings of Fan and Wang (2013). Moreover, our results suggest that organizational identification significantly strengthens intention to stay, a finding that supports those of Bharadwaj and Yameen (2020) and Kim et al. (2020). This signified that during the pandemic, although employees worried about their job prospects and perceived all industries to be experiencing an economic slump, they were willing to stay in their job and that organizational identification plays a crucial rule in strengthening hospitality industry employees' intention to stay during such economic crisis.
In summary, our results were consistent with the predictions of the theoretical model. The significant positive effect of job insecurity on organizational identification observed contradicts the findings of Piccoli et al. (2017) and Callea et al. (2016). Regarding job insecurity and intention to stay, our findings are consistent with those of Fan and Wang (2013). Organizational identification and intention to stay accords with the findings of Bharadwaj and Yameen (2020) and Kim et al. (2020).

Recommendations
How employees thought during the pandemic differed from that at other times. However, a company's management and environment are truly revealed only at times of crisis. Among the items used in this study to assess job insecurity, the highest mean was obtained for "My working hours may become irregular because of the pandemic," showing that when hospitality companies decrease employee working hours, the employees are likely to perceive job insecurity. Thus, companies should plan different work schedules for off seasons, peak seasons, and pandemic times and announce such schedules during educational training to enable employees to prepare and make relevant arrangements in advance. Among the items used to assess organizational identification and intention to stay, only "I care about the future of my company" achieved a high score, revealing that the hospitality industry is an industry with a high turnover rate and indicating that hospitality companies should enhance their recruitment and training operations to respond to various situations.
This study is one of few to document the significant pairwise relationship among job insecurity, organizational identification, and intention to stay by using path analyses. However, the study entailed several limitations. Cultural practices and epidemic prevention measures differ among countries, industries, and employers. In addition, the durations of outbreaks and severity of the pandemic's impact vary by country, so does the impact on the hospitality industry. Accordingly, the influence of the duration of COVID-19 outbreak on the hospitality industry warrants further investigation. Moreover, differences in employee position or seniority may result in distinct effects on their intention to stay, job insecurity, and organizational identification. A key policy implication derived from the current findings is that establishing favorable relationships with employees and promoting their organizational identification can help strengthen their intention to maintain their present organizational role.