How Does a Job Insecurity Climate Emerge in an Organization and How is it Associated with Organizational Antecedents?
Job insecurity has attracted growing scientific interest, especially now due to the economic slowdown in Europe. This interest is based on the observed increase in job instability in the labor market and the higher risk of job loss for employees due to, for example, continuous downsizing, organizational restructuring, the growth of temporary employment and the current economic crisis (De Cuyper et al. 2012). Indeed, interest in this topic was particularly prevalent in the 2000s (i.e., Greenhalgh & Rosenblatt 2010; Sverke et al. 2010).
Most of this literature has defined job insecurity from an individual perspective, conceptualizing it as an individual’s concern about their own job continuity in the future (De Witte 1999). Therefore, by definition, job insecurity is a subjective perception, based on the individual’s experiences of job/employment instability. However, job insecurity also arises within an organizational context, defined by human resource practices and employment policies. This suggests that employees in the same workplace may share similar perceptions of job insecurity (De Cuyper et al. 2009; Sora et al. 2023). The idea that job insecurity is shared can be explained using several theoretical frameworks. For example, the Social Information Processing Model (Salancik & Pfeffer 1978) explains shared perceptions by arguing that individuals use information about their own past experiences as well as others’ opinions to form their own perceptions. Socialization Theory (Schneider & Reichers 1983), in turn, explains how people who have worked for a long time in an organization transmit their perceptions, values, and norms to newcomers, so that newcomers can assimilate them. Similarly, Social Comparison Theory (Festinger 1954) highlights that individuals interact with referent others to understand and make sense of uncertain situations, such as job insecurity. All these theoretical frameworks support the view that shared perceptions emerge as a property of a social unit – such as a team within the same workplace – through social interaction, exchange, and amplification (Kozlowski & Klein 2000). Due to this theoretical reasoning and calls to study job insecurity from a multilevel perspective (i.e., Mohr 2000; Sora et al. 2009; Sora et al. 2023), limited but growing research has examined the emergence of shared perceptions of job insecurity. Job insecurity climate has been conceptualized according to two composition models proposed by Chan (1998): the direct consensus model (e.g., Sora et al. 2023) and the referent-shift consensus model (e.g., Låstad et al. 2016). The main difference between them lies in the referent used for assessing consensus – employees in the direct consensus model and the organization in the referent-shift consensus model. In the direct consensus model, employees report their own experiences (e.g., ‘I feel insecure about my job’), and these responses are aggregated to indicate the organization’s climate. In contrast, the referent-shift consensus model asks employees to rate their perception of the group as a whole (e.g., ‘Employees in this organization feel insecure about their jobs’), and these group-focused ratings are also aggregated to form an organizational climate measure. Both conceptualizations are considered measures of organizational climate. In this study, we adopted the direct consensus model because this is the main conceptualization used in the general literature on organizational climate and researchers have found that the direct consensus model can be a stronger predictor of job attitudes than the referent-shift consensus model (e.g., Wallace et al. 2016). Nevertheless, the literature seems to confirm that individual perceptions of job insecurity are shared by members of the organization to form a job insecurity climate at the organizational level, which affects individual outcomes beyond the impact of job insecurity at the individual level (De Cuyper et al. 2009; Sora et al. 2009; Sora et al. 2012). Sora et al. (2012, p. 385) defined the job insecurity climate as ‘a shared concern about the continued existence of the job’. Therefore, this climate is regarded as a threatening characteristic of the organizational environment and, in line with appraisal theory (Lazarus & Folkman 1984), can function as a stressor.
Despite this relevant insight, most of the literature on job insecurity has adopted an individual perspective without considering the shared perception of job insecurity. Therefore, the present study aims to contribute to the literature by examining job insecurity as a collective phenomenon. Furthermore, to the best of our knowledge, there is a notable absence of research on the antecedents of job insecurity climate, in contrast to research focused on its outcomes. We are only aware of the study by De Cuyper et al. (2009), which showed how the extensive use of temporary employment created a climate of job insecurity among permanent workers. Accordingly, there is a critical lack of knowledge about the antecedents of job insecurity climate. Therefore, the main contribution of this study was to identify the critical organizational characteristics and processes that may determine the job insecurity climate within an organization.
We developed our hypotheses in relation to the existing literature on job insecurity at the individual level and previous research on organizational climate. As previously mentioned, the literature on job insecurity suggests that having a high proportion of temporary workers within an organization can affect the job insecurity climate of permanent employees (De Cuyper et al. 2009). In addition, Sora et al. (2012) showed that the job insecurity climate and its effects differed between countries, which they highlighted by comparing Spain and Belgium. It is also worth noting that the literature on organizational climate has demonstrated that group-level factors are among the strongest antecedents for explaining the organizational climate and its subtypes (Kuenzi & Schminke 2009). Among these group-level antecedents, the literature highlights the type of organization as a significant factor in shaping the organizational climate, as sector characteristics (e.g., public vs. private) create distinct environments that influence its formation (Ingrams 2018). Organizational resilience is increasingly recognized as a critical antecedent of organizational climate, influencing how organizations respond to crises and adapt to changing environments (Ayoko 2021). In addition, another important antecedent of the organizational climate is communication. Effective communication is critical for enhancing a positive climate (Oo & Wai 2020; Torres 2022). In this vein, rumors, as a form of informal communication, can become a primary source of information in uncertain situations and, thus, can shape the organizational climate (Bünyamin 2021; Şafak & Kahveci 2021).
Given that the literature on organizational climate and job insecurity climate represents work environments characterized by an overall concern about job stability, it seems plausible to propose these organizational factors as determinants of the job insecurity climate. This study aims to examine whether the different characteristics and processes at play in organizations, such as their type (private vs. public), the percentage of temporary workers in the organization, organizational resilience, the quality of communication and the existence of rumours, are related to the job insecurity climate. In addition, this study also examines how a contextual factor such as the country can affect the job insecurity climate. Our analysis is based on data from two different sources: employees and employers in two European countries (Spain and Austria).
Organization type
The organization may be public or private, and its type can present different characteristics:
Ownership. Public organizations are controlled predominantly by political forces, whereas private organizations are controlled by market forces (Clarkson 1972).
Purpose and goals. Private organizations usually pursue the single goal of profit, i.e. ‘their success – or failure – in the market’ (Farnham & Horton 1996). In contrast, public organizations aim to serve society by providing essential services, promoting social welfare, and ensuring the equitable distribution of resources (Boyne 2002).
Decision-making process. Public organizations usually have more formal decision-making procedures and are less flexible compared to private organizations in general (Bozeman & Kingsley 1998).
Job security. In public sector organizations, HRM policies and practices related to personnel issues are typically less flexible, as they operate under principles of public governance and accountability and are regulated by specific rules on hiring, dismissal, and promotion (Boyne 2002). Procedures for taking larger actions are complex and time-consuming, to the extent that they are not usually applied, and public employees enjoy greater job security (Baldwin 1987).
This suggests that while job insecurity may act as a general stressor for all employees, its severity can vary according to the type of organization in which they work. Higher job insecurity seems to be present in private organizations because they are more strongly influenced by market variation and economic downturns during recessions or periods of financial instability. Moreover, private organizations tend to have greater flexibility in decision-making when adopting downsizing measures. In contrast, public organizations operate in the public interest and adhere to principles of public governance and accountability. Thus, their formal procedures make it more difficult to adopt downsizing measures. Therefore, studies consistently show that public-sector employees enjoy a higher level of job security (Baldwin 1987; Erlinghagen 2008; Khojasteh 1993; Maidani 1991; Mauno & Kinnunen 2002; Noori 2023; Ruvio & Rosenblatt 1999).
However, this literature examined levels of job insecurity at an individual level, even though the type of organization constitutes an organizational characteristic that could be considered a contextual factor in both research design and analysis. Following the theories and literature outlined above, we propose that private organizations have higher levels of job insecurity climate compared to public organizations. Accordingly, we hypothesize that:
Hypothesis 1. Private organizations have higher levels of job insecurity climate than public organizations.
Percentage of temporary workers in organizations
Employment protection legislation has been modified in most industrialized countries over recent decades due to neoliberal policies.
These changes aim to achieve greater numerical flexibility in the workforce, enabling organizations to better adapt to ongoing economic transformations (e.g., globalization, new technologies, downsizing, etc.), improve their competitiveness and contribute to a more dynamic labor market (OECD 2004). Implementing these new regulations has provided employers with an opportunity to circumvent strict regulations imposed on permanent employment, increase the numerical flexibility of their organizations and cut costs. Hence, the use of this new form of employment has become widespread (OECD 2009).
Differences in the strictness of the employment protection legislation for regular and temporary employment can aggravate labor market duality (OECD 2004). The structure of the labor market in general, and the workforces within organizations more specifically, can show core-periphery differentiation (Segmentation Theory or Core-Periphery Model; Atkinson 1984). Temporary employees are at the periphery of the labor market, whereas permanent workers form its core. The peripheral workforce is contracted to meet the organization’s fluctuating needs and is often assigned to less desirable and more unstable jobs, with fewer resources. These roles are typically designed to be easily transferable between workers, requiring minimal investment from employers (e.g., in training). In contrast, the core labor force consists of employees regarded as essential to the organization and, as such, they typically occupy the most stable, well-resourced, and desirable positions. Therefore, temporary employees are more vulnerable to job insecurity due to perceiving themselves as more easily dispensable in the organization compared to permanent employees (Kinnunen & Nätti 1994; Peiró, Sora & Caballer, 2012; Sverke et al. 2000).
However, permanent employees are not exempt from the implications of this employment protection legislation, and especially from the promotion of temporary employment. An organization may use temporary employment to supplement permanent workers; thus, this strategy is intended to stabilize and secure permanent positions. However, using temporary employment can also be interpreted as a way to reduce employers’ obligations when compared to long-term employment (Stanworth & Druker 2006). According to Social Identity Theory (Tajfel & Turner 1986), low status groups could be interpreted as a threat to the position of high-status groups, especially when the low-status group begins to become dominant. In combination with the previously mentioned Segmentation Theory, temporary workers constitute a low-status peripheral workforce, whereas permanent workers are in the core and enjoy a high status. Furthermore, with an increase in the temporary workforce in the organization, group boundaries may become more blurred (Pearce 1998), especially when it comes to permeable boundaries, such as temporary employment that can be transformed into permanent jobs (Scherer 2004; von Hippel 2006). This also has negative implications for permanent workers, such as status loss (Barnett, W P & Miner, 1992; Geary 1992) and an increase in their responsibilities and workload (Geary 1992; Pearce 1993). Another implication for permanent workers, which has received less attention, is the increase in perceptions of job insecurity (Geary 1992). Indeed, De Cuyper et al. (2009) found an association between the proportion of temporary workers in the total workforce of an organization and the job insecurity climate of permanent workers in the same organization.
In line with this evidence, we conclude that the extent of temporary employment in an organization is associated with the level of the job insecurity climate among workers independently of their individual type of contract. Thus, we hypothesize:
Hypothesis 2. The percentage of temporary employees in the organization is positively associated with the level of job insecurity climate.
Organizational resilience
Organizational resilience has been studied in the literature on disaster management, crisis management, and high-reliability organizations. Nevertheless, there is no consensus regarding its definition or the factors that constitute it. Organizational resilience has mainly been understood as a set of dynamic and flexible organizational capabilities that are used to manage unexpected and stressful events, such as predictive, survival, adaptive, coping, and learning capabilities (Chen, Ya-ping & Liu 2021). Thus, organizational resilience reflects an organization’s ability to cope with and recover from unexpected and adverse stressful situations and pick up where they left off (Munoz, Billsberry & Ambrosini 2022). Organizational efforts are related to coping strategies and a quick ability to resume expected performance levels (Rudolph & Repenning 2002; Sutcliffe & Vogus 2003). It is an important factor that enables an organization to leverage its resources and capabilities, not only to resolve current dilemmas, but to benefit from opportunities and build a successful future (Coutu 2002; Lengnick-Hall & Beck 2005; Solá & Sora 2007; Weick 1988). Thus, organizational resilience is more than mere survival; it includes the ability to identify potential risks, take proactive steps (Longstaff 2005) and ensure that the organization thrives in the face of adversity (Kanigel 2001). Indeed, in the literature, organizational resilience has been related to several beneficial outcomes, such as competitiveness, survival, performance, and safety (Hamel & Valikangas 2003; Hollnagel et al. 2006; i.e. Bustinza et al. 2019; West, Dyrbye & Sloan 2009).
Given that resilience is an essential positive capacity for organizations, one that allows them to recover and rebound when they are faced with stressful situations, members of resilient organizations should also perceive these events as less threatening than members of organizations characterized by lower resilience (West, Dyrbye & Sloan 2009). Following this reasoning, it seems plausible to suggest that employees in resilient organizations feel more secure about their job continuity because they share the belief that their organization can cope with difficult situations, will succeed, and consequently, lay-offs might be less likely. In fact, several studies have empirically demonstrated the relationship between organizational resilience and lower job insecurity at the individual level (Prasongthan 2022; Vo-Thanh et al. 2022). However, we are not aware of any study that has examined this relationship at the organizational level. Thus, we hypothesize that:
Hypothesis 3. Organizational resilience is negatively associated with the level of job insecurity climate.
Communication and rumors
The quality of intra-organizational communication is a crucial aspect of an organization and is even considered its lifeblood (Allen & Griffeth 1997). More specifically, intra-organizational communication is understood as ‘the degree to which information is transmitted among members of an organization’ (Kyoungjo, Youngbae & Jinjoo 1991, p. 17). Organizations are open systems that frequently need to process uncertain and equivocal information (March & Simon 1958). According to uncertainty theory, uncertainty reflects an ambiguous and unpredictable situation where information is unavailable or inconsistent (Babrow 2001). This uncertainty is an aversive psychological state and motivates strategies to reduce it, such as looking for information (Ashford & Black 1996; Bordia et al. 2006; Brashers 2001). However, adequate communication may reduce the uncertainty and discomfort that are common in relation to job insecurity and may even reduce insecurity. For example, Robbins (1999) suggested that adequate organizational communication – providing early, open, realistic, and transparent information – positively influences employees’ reactions during downsizing or mergers. Such communication helps employees understand the meaning of events, take appropriate actions, and thereby promote predictability while reducing uncertainty. Charoensukmongkol and Suthatorn (2022) demonstrated how high-quality vertical communication, characterized by trustworthiness, candidness, reciprocity, information exchange, feedback, tolerance for differing opinions and negotiation, were able to reduce the levels of perceived job insecurity of flight attendants during the COVID-19 pandemic in 2019. Similarly, Saeed et al. (2023) showed how effective communication reduced job insecurity levels in a sample composed of salespersons from pharmaceutical companies in Pakistan. This assumption can be extrapolated to the study of the job insecurity climate; therefore, we hypothesize:
Hypothesis 4: Communication is negatively associated with the level of job insecurity climate.
However, when the information is neither adequate nor inadequate, but simply not available from formal or trusted sources, people tend to resort to informal problem solving. Rumors are a product of this collective sense-making process, defined as ‘unverified statements regarding issues of topical significance’ (Bordia et al. 2006, p. 602). Through rumors, employees try to restore a feeling of understanding by giving meaning to the uncertain situation (DiFonzo & Bordia 1997; Shibutani 1966; Smet et al. 2016). Nevertheless, rumors only spread when there is a need to reduce uncertainty; and this only concerns matters of particular relevance or importance for organizational members. Indeed, Allport and Postman (1947) suggested that rumor activity is directly proportional to the ambiguity of an issue multiplied by its importance. Rumors in the workplace tend to be about issues such as job insecurity or work conditions (DiFonzo, Bordia & Rosnow 1994).
Rumors can provoke stress (Bordia et al. 2006; DiFonzo, Bordia & Rosnow, 1994; DiFonzo & Bordia 2000) because anticipating a probable job loss involves anxiety, discomfort and even panic reactions among those who hear and believe these rumors. For example, during organizational change, Smollan (2015) showed that employees who had heard rumors before the formal announcement tended to experience more uncertainty and a greater degree of emotional distress. This is congruent with cognitive dissonance theory (Festinger 1957), which argues that rumors often justify and give meaning to existing anxiety. As Bordia et al. (2006) exemplified, organizational changes are associated with higher levels of anxiety among employees because of the uncertainty and loss of control. In addition, this anxiety was also amplified by rumors suggesting future job loss.
In fact, rumors can create confusion, increasing employees’ job insecurity levels (Smet et al. 2016). Hence, it seems plausible that rumors about potential job loss could be associated with an overall concern about job continuity. Thus, we hypothesize:
Hypothesis 5. Rumors are positively associated with the level of job insecurity climate.
Differences among European countries
European countries have different characteristics that are critical for job instability and job insecurity levels. For example, in countries with strong, stable, and growing economies, the workforce will generally experience lower levels of job insecurity because they perceive that there are more employment opportunities. In this study, we focused on two European countries: Spain and Austria. These two countries can be seen as prototypical examples of European countries with relatively unstable versus relatively stable labor markets. From 2011 to 2021, official labor market data showed higher mean rates of temporary employment (25.1%), and unemployment (19.4%) in the Spanish labor market compared to the Austrian labor market, which presented lower temporary employment (9%) and unemployment rates (5.3%) (Eurostat 2023). Accordingly, Spanish employees reported higher levels of job insecurity and job insecurity climate compared to Austrian employees (Sora et al. 2023). Furthermore, social welfare can affect job insecurity, depending on its typology (Esping-Andersen 1990). Austria is in the conservative-continental cluster of welfare states, which are characterized by a medium level of social services and medium status-related services. In contrast, Spain, as a Mediterranean country, is in a subcategory of the conservative-continental welfare states, which have a weaker and fragmented system of social protection and higher acceptance of status differences. Thus, Spanish employees have a poorer social welfare net and, therefore, experience higher job insecurity levels compared to Austrian employees. Finally, the institutional and social structure may also be determinants in the job insecurity levels of a country. Erlinghagen (2008) found significant cross-national variation in self-perceived job insecurity in 17 European countries. This author suggested that this variability was due to institutional and social structural differences. König et al. (2011) showed that the relationship between job insecurity and job attitudes was more negative in the US than in Switzerland. They proposed that social safety nets could explain these results. The Swiss social safety net is stronger than the social safety net in the US in several aspects, such as eligibility criteria, guaranteed duration of benefits, level of unemployment compensation, and post-layoff continuance of health care coverage. In this respect, the International Labour Organization (2004) classified Switzerland as exemplary in terms of its policies to protect workers compared to the US system, which offers far fewer economic security protections.
To conclude, according to the previous literature, job insecurity is present at different levels from country to country. Considering that these characteristics can promote higher job instability and job insecurity, we hypothesize that:
Hypothesis 6. Spain shows a higher level of job insecurity climate than Austria.
Method
Procedure
Data for the present study were collected in two European countries: Spain and Austria. This study is part of a broader research project examining the job insecurity climate in different European countries in 2011. To obtain the sample we used a convenience sampling strategy with specific criteria that organizations had to meet to collaborate in the research: 1. the organization must be in one of the following labor sectors: health, construction, education or retail; 2. the organization must have permanent and temporary employees in order to have participants with different types of contracts; and 3. the organizations must have a minimum size of 3–4 employees due to multilevel methodological requirements. The sampling strategy consisted of the researchers contacting the human resources managers or directors of each organization to explain the purpose of the research and to request their collaboration. Once they had agreed to participate, two different questionnaires were distributed within the organization: one for human resource managers or the main manager, and another for employees. The researchers made an appointment to collect the completed questionnaires at a later time (approximately one week). Participation was voluntary and anonymity and confidentiality were guaranteed. Thus, in each organization, we collected a manager’s questionnaire, which included information about the organization, and the questionnaires of the employees who had agreed to participate voluntarily, with data about their perceptions and attitudes. Given that we had to rely on voluntary participation, the sampling method could not be completely random. The response rate ranged from 7.6% to 100%, with a mean of 59.38%. The study was conducted in accordance with the Declaration of Helsinki, and the protocol followed the guidelines of the Ethics Committee of our university (code: CEIPSA-2025-TDO-0091).
Sample
The sample consisted of Spanish and Austrian respondents. Specifically, it was made up of 1,466 employees from 141 organizations. Overall, 65% of employees (n = 958) and 64% of organizations (n = 91) were Spanish, whereas 35% of employees (n = 508) and 36% of organizations (n = 50) were Austrian. This sample was collected across four labor sectors in the two countries: the construction industry (norganization = 26; nemployees = 220), the retail industry (norganization = 47; nemployees = 443), health (norganization = 27; nemployees = 331), education (norganization = 38; nemployees = 441), and others (norganization = 3; nemployees = 31). Organizational size ranged from 3 to 3,000 employees (M = 111.60, SD = 334.42), and the organizational size of entities included in the study was from 3 to 62 (M = 10.39, SD = 10.52). Overall, 919 employees were women (62.7%), and 507 were men (34.6%), with a mean age of 38.27 years (SD = 10.75). In total, 77.9% of employees had permanent contracts (n = 1142), and 20.7% had temporary contracts (n = 304). Finally, 150 employees had completed primary education (10.2%), 699 employees had completed secondary education (47.7%), 489 had university studies (33.4%), 39 had post-graduate studies (2.7%), 36 had other studies (2.5%), and one employee had no studies (0.1%).
Measures
The study data came from two different sources: managers and employees. Managers provided information about their labor sector, the percentage of temporary employment in the organization, the organizational size, and the type of the organization. Employees responded to questions about the job insecurity climate, organizational resilience, communication, and rumors.
Organizational size was determined with the following question: ‘How many employees are working for this organization, including those with temporary contracts or contracts with another company, such as agency workers or subcontractors?’.
Labor sectors were coded as (1) retail, (2) education, (3) construction, and (4) health.
Country was coded by the researchers into 1 (Spain) and 0 (Austria).
Type of organization was measured as a dummy variable with two response options: 0 = public and 1 = private. Participants were asked to respond to the statement: ‘Your organization is…”.
Percentage of temporary employees in the organization was calculated as the proportion of temporary workers. This was derived from the total number of workers (including those with temporary contracts; ‘How many employees are working for this organization, including those with temporary contracts or contracts with another company, such as agency workers or subcontractors?’) and the total number of permanent workers (‘How many permanent employees are working for this organization?’).
The job insecurity climate was measured by aggregating employees’ perceptions. Job insecurity was assessed with a 4-item scale (Vander Elst et al. 2014). An example of the items is: ‘Chances are, I will soon lose my job’. Responses ranged from 1 (strongly disagree) to 5 (strongly agree). The Cronbach’s alpha for this scale was 0.88.
Organizational resilience was assessed with a 10-item scale by Khanbeiki et al. (2025). One example of an item is: ‘Members of this organization used to manage difficulties at work’. The response range was from 1 (strongly disagree) to 5 (strongly agree). The Cronbach’s alpha was 0.80.
Communication was measured using the following question: ‘To what extent is the information you receive from your organization usually:’, followed by four items: 1. You get it when you need it (timely); 2. You can rely on it (accurate); 3. You can use it (useful); 4. You can understand it (clear) (Allen, & Brady 1997). The response range was from 1 (strongly disagree) to 5 (strongly agree). The Cronbach’s alpha was 0.91.
Rumors. Based on Bordia et al. (2006), we developed a scale with the statement ‘How often have you heard the following rumours in your organization?’ and the following six items: (1) Downsizing and job losses; (2) Increased workload for less pay; (3) Lack of career advancement; (4) Loss of work or job-related facilities; (5) Changes to the structure and type of the organization (i.e. work unit closures, mergers or partial privatization of the organization); and (6) Statements alleging staff turnover. The response range was from 1 (never) to 5 (very often). The Cronbach’s alpha was 0.88.
Analysis
Firstly, we examined the validity of our measures in the two countries. Multi-group confirmatory factor analysis (MCFA) was used to examine the validity of the measures: job insecurity, organizational resilience, organizational communication, and rumors. The global model fit was determined through different indexes (Marsh, Balla & Hau 1996): Chi-square index (χ2), chi-square/df ratio (χ2/df), Normed Fit Index (NFI), Comparative Fit Index (CFI), Root Mean Square Error of Approximation (RMSEA), and Root Mean Square Residual (RMR). In addition, measurement invariance was computed to examine whether there was measurement equivalence in our cross-cultural sample (Byrne & Van de Vijver 2010). Hence, three models were tested using multigroup CFA, which progressively fixed the number of items and factors (configural invariance -proposed two-factor model), the factor loads (metric invariance), and the intercepts (scalar invariance). The tested model should not present differences in Chi-square (ΔChi-squared), CFI (ΔCFI) < 0.01, McDonald’s NCI (ΔMcDonald’s NCI) < 0.02, or RMSEA (ΔRMSEA) < 0.01 when compared to other models. These analyses were computed with the AMOS version 29 software.
Secondly, the objective of this study was oriented towards the organizational level. Following the aggregate properties model (Chen, Mathieu & Bliese 2004), the level of the construct, measurement, and use in substantive analyses were aligned at the same level of analysis (organizational level) for organizational size, labor sector, and percentage of temporary employment. In contrast, organizational resilience, organizational communication, and rumors were conceptualized according to the referent-shift consensus model (Chan 1998). The level of the constructs, their measurements, and their use in substantive analyses were not aligned. There was a shift in the referent in the measure to represent the higher-level construct. In other words, the referent in the measures referred to the organization. The analyses were also at the organizational level, but the data were collected at the individual level. Hence, the lower-level constructs (individual level) were adapted to the higher-level construct through within-group consensus. Therefore, individual scores on job insecurity, organizational resilience, communication, and rumors provided by employees had to be aggregated at the organizational level. Before aggregation, a set of requisites had to be met, including within-organizational agreement and between-organizational variance. To determine the within-organizational agreement, rwg(j) (James, Demaree & Wolf 1993) and the average deviation index (ADM(J)) (Burke, Finkelstein & Dusig, 1999; Burke & Dunlap 2002) were computed. The between-organizational variance was computed through a one-way analysis of variance (ANOVA). Finally, Bliese (2000) recommended also verifying the reliability of the measure by means of the intraclass correlation indexes (ICC1 and ICC2). ICC1 presents the proportion of total variance that can be explained by group membership, and ICC2 indicates the degree to which group means can be reliably differentiated.
Thirdly, descriptive statistics and correlation analyses were computed. Lastly, the hypotheses were tested through a multiple regression analysis with job insecurity climate as the dependent variable. This multiple hierarchical regression was computed at the organizational level (n = 141). As described by Cohen and Cohen (1983), in step 1, organizational size and labor sector were introduced as control variables. The predictor variables were entered in separate steps, allowing the explained variance of each predictor variable to be obtained (these being organization type, percentage of temporary employment in the organization, organizational resilience, communication, rumors and country). Finally, we used grand-mean centered scores to solve the possible problem of multicollinearity (see Hofmann & Gavin 1998). The data were analyzed using the SPSS software program, Version 29.
Results
The results of the multi-group confirmatory factor analysis indicated an acceptable model fit for the four measured constructs: job insecurity, organizational resilience, communication and rumors (see Table 1). The Chi-Squared Goodness-of-Fit Index (χ2) did not show a satisfactory fit between the observed covariance matrix and the hypothesized models. This was probably due to the sensitivity of this index to large sample sizes. However, NFI, IFI and CFI values were higher than the cutoff of 0.90 (Jöreskog & Sörbom 1993) and RMSEA and RMR were lower than 0.08 (Browne & Cudek 1993).
Table 1
Multi-group confirmatory factor analysis: Job insecurity, organizational resilience, communication, and rumors.
| χ2 | df | p-VALUE | χ2/df | NFI | IFI | CFI | RMR | RMSEA | |
|---|---|---|---|---|---|---|---|---|---|
| MCFA: Job insecurity | 12.15 | 4 | 0.01 | 3.03 | 0.99 | 0.99 | 0.99 | 0.012 | 0.040 |
| MCFA: Organizational resilience | 384.21 | 68 | 0.00 | 5.65 | 0.90 | 0.91 | 0.91 | 0.045 | 0.060 |
| MCFA: Communication | 105.55 | 4 | 0.00 | 26.38 | 0.97 | 0.97 | 0.97 | 0.026 | 0.139 |
| MCFA: Rumors | 183.32 | 18 | 0.00 | 10.18 | 0.95 | 0.96 | 0.96 | 0.069 | 0.084 |
We conducted measurement invariance testing to ensure that the constructs had equivalent meaning across both national groups in the sample (see Table 2). Overall, the results indicated that there were no differences for factor loadings between the countries when configural invariance and the metric invariance models were compared for all our measures because Δchi-square revealed no significant differences, ΔMc NCI was lower than 0.02, and ΔCFI and ΔRMSEA were lower than 0.01. It is noteworthy that ΔRMSEA for job insecurity and communication and ΔCFI for rumors were slightly higher, with values of 0.02. However, according to Vandenberg and Lance (2000) and Cheung and Rensvold (2002), a slightly higher value, such as 0.02, may still be acceptable, especially if the other fit indicators are favorable. Likewise, no differences between countries were found for item intercepts in the communication and rumors measures when the metric invariance and the scalar invariance models were compared. All indexes were lower than their respective cutoffs. However, for job insecurity and organizational resilience, differences were found for item intercepts because ΔCFI values were higher than 0.01 (0.03 for job insecurity and 0.05 for organizational resilience), ΔMc NCI was 0.05 for organizational resilience, and ΔRMSEA was 0.03 for job insecurity. Thus, the intercepts of job insecurity and organizational resilience are partially invariant. Hence, following Byrne and Van de Vijver (2010), partial scalar invariance was examined by releasing item 2 in job insecurity and items 3, 5, 6, and 7 in organizational resilience. The results presented no differences in the comparison between metric invariance and the partial scalar invariance models for item intercepts. This suggests that measurement invariance was supported for both constructs.
Table 2
Invariance measurement.
| Δχ2 | Δdf | p-VALUE | ΔCFI | ΔMc NCI | ΔRMSEA | |
|---|---|---|---|---|---|---|
| Job insecurity: Metric invariance | 30.13 | 3 | 0.00 | 0.00 | 0.01 | 0.02 |
| Job insecurity: Scalar invariance | 115.78 | 10 | 0.00 | 0.03 | 0.02 | 0.03 |
| Job insecurity: Partial scalar invariance | 69.42 | 9 | 0.00 | 0.01 | 0.00 | 0.01 |
| Organizational resilience: Metric variance | 73.66 | 9 | 0.00 | 0.01 | 0.02 | 0.00 |
| Organizational resilience: Scalar variance | 190.68 | 9 | 0.00 | 0.05 | 0.05 | 0.01 |
| Organizational resilience: Partial scalar variance | 32.18 | 5 | 0.00 | 0.01 | 0.00 | 0.00 |
| Communication: Metric invariance | 6.12 | 3 | 0.10 | 0.00 | 0.00 | 0.02 |
| Communication: Scalar invariance | 20.49 | 3 | 0.00 | 0.00 | 0.00 | 0.01 |
| Rumors: Metric invariance | 49.30 | 5 | 0.00 | 0.02 | 0.01 | 0.00 |
| Rumors: Scalar invariance | 72.48 | 5 | 0.00 | 0.01 | 0.02 | 0.01 |
Once our measures had been validated across countries, the consistent pattern of results for the previous prerequisites justified the aggregation of all variables reported by employees in the organizations. The Average Deviation Index (ADM(J)) and the rwg(j) supported within-organization agreement for job insecurity (ADM(J) = 0.68 (median = 0.68); rwg(j) = 0.74 (median = 0.84), organizational resilience (ADM(J) = 0.63. (median = 0.64); rwg(j) = 0.93 (median = 0.94); communication (ADM(J) = 0.63 (median = 0.68); rwg(j) = 0.83 (median = 0.88)); and rumors (ADM(J) = 0.73 (median = 0.76); rwg(j) = 0.74 (median = 0.85). All these values indicated within-organization agreement for all variables as ADM(j) values were lower than the recommended value of 0.83 (Burke, Finkelstein & Dusig 1999; Burke & Dunlap 2002) and the values of rwg(j) were higher than the critical value of 0.70 (James, Demaree & Wolf, 1993). The analysis of variance (ANOVA) confirmed significant between-organization differences for all variables: job insecurity (F(139 1306) = 5.52, p < 0.01); organizational resilience (F(139 1315) = 4.32, p < 0.01); communication (F(139 1308) = 5.59, p < 0.01); and rumors (F(139 1308) = 11.28, p < 0.01). Finally, the inter-rater reliability index (ICC1) and the reliability of group mean index (ICC2) provided additional evidence for valid constructs at the organizational level: job insecurity (ICC1 = 0.30; ICC2 = 0.81); organizational resilience (ICC1 = 0.24; ICC2 = 0.77); communication (ICC1 = 0.31; ICC2 = 0.82); and rumors (ICC1 = 0.50; ICC2 = 0.91).
Descriptive statistics (means, standard deviations) and inter-correlations are presented in Table 3. Most correlations were significant yet moderate, with a range from –0.18 to 0.64. Specifically, correlations between organizational resilience, communication and rumors were moderately high.
Table 3
Descriptive statistics (means and standard deviations) and correlations.
| MEAN | SD | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | |
|---|---|---|---|---|---|---|---|---|---|---|
| 1. Country (dummy) | – | – | – | |||||||
| 2. Organizational size | 1110.6 | 334.42 | –0.07 | – | ||||||
| 3. Job Insecurity Climate | 2.03 | 0.65 | 0.41** | 0.10 | – | |||||
| 4. Organization type (dummy) | – | – | 0.13 | 0.03 | 0.23** | – | ||||
| 5. % temporary workers | 19.66 | 97.82 | 0.12 | 0.22* | 0.27** | 0.09 | – | |||
| 6. Organizational resilience | 3.51 | 0.35 | –0.18* | 0.00 | –0.32** | 0.04 | –0.24** | – | ||
| 7. Communication | 3.72 | 0.59 | –0.05 | –0.12 | –0.52** | –0.08 | –0.32** | 0.55** | – | |
| 8. Rumours | 2.42 | 0.85 | 0.31** | 0.06 | 0.64** | –0.21* | 0.26** | –0.36** | –0.54** | – |
[i] Note. Note: n = 141 organizations. * p < .05; ** p < .01. Organization type and Country are dummy variables (0 = public, 1 = private; 0 = Austria, 1 = Spain).
Table 4 presents the results of the multiple hierarchical regression predicting the job insecurity climate, after controlling for organizational size and labor sector.
Table 4
Summary of hierarchical regression analyses: Predictors of job insecurity climate.
| ESTIMATE | SE | 95% CI | P | R2 | ΔR2 | ||
|---|---|---|---|---|---|---|---|
| LL | UL | ||||||
| Step 1 | 0.079 | 0.079 | |||||
| Organizational size | 0.068 | 0.000 | 0.000 | 0.000 | 0.462 | ||
| Retail sector | 0.161 | 0.375 | –0.527 | 0.960 | 0.565 | ||
| Education sector | 0.069 | 0.378 | –0.651 | 0.846 | 0.797 | ||
| Construction sector | 0.353 | 0.383 | –0.204 | 1.314 | 0.150 | ||
| Health sector | 0.060 | 0.389 | –0.668 | 0.873 | 0.792 | ||
| Step 2 | 0.120 | 0.041 | |||||
| Organization type | 0.275* | 0.183 | 0.052 | 0.776 | 0.025 | ||
| Step 3 | 0.189 | 0.069 | |||||
| % Temporary workers in the organization | 0.273** | 0.001 | 0.001 | 0.003 | 0.003 | ||
| Step 4 | 0.252 | 0.063 | |||||
| Organizational resilience | –0.267** | 0.163 | –0.817 | –0.0169 | 0.003 | ||
| Step 5 | 0.345 | 0.093 | |||||
| Communication | –.385** | 0.106 | –0.622 | –0.203 | 0.001 | ||
| Step 6 | 0.617 | 0.271 | |||||
| Rumours | 0.659*** | 0.056 | 0.377 | 0.601 | 0.001 | ||
| Step 7 | 0.644 | 0.027 | |||||
| Country | 0.194** | 0.095 | 0.079 | 0.455 | 0.006 | ||
[i] Note. n = 141 organizations. * p < 0.05; ** p < 0.01. *** p < 0.001. Estimate refers to unstandardized coefficients. Organization type, Labour sector and Country are dummy variables (0 = public, 1 = private; 0 = Austria, 1 = Spain).
Hypothesis 1 proposed a relationship between organization type (private vs. public) and job insecurity climate. Private organizations reported a higher job insecurity climate than public organizations (B = 0.27, p = 0.02; 95% Confidence Interval [CI] 0.05, 0.77). The results supported hypothesis 1. Hypothesis 2 was also confirmed. As predicted, the percentage of temporary workers in the organization was positively related to the job insecurity climate; thus, the more temporary staff, the more job insecurity (B = 0.27, p = 0.0; 095% CI 0.00, 0.00).
Hypotheses 3 and 4 predicted a negative relationship between organizational resilience and communication and the job insecurity climate, respectively. The results supported these hypotheses; higher levels of organizational resilience (B = –0.26, p = 0.00; 95% CI –0.62, –0.20) and good communication (B = –0.38, p = 0.00; 95% CI –0.81, –0.16) were associated with a lower job insecurity climate. Hypothesis 5 concerned the relationship between rumors and the job insecurity climate. The results confirmed this hypothesis as well. A higher frequency of rumors related to organizational changes was significantly associated with a higher job insecurity climate within organizations (B = 0.65, p = 0.00; 95% CI 0.37, 0.60). Finally, the country variable contributed significantly to predicting the job insecurity climate. Spanish organizations showed higher job insecurity climates compared to Austrian organizations (B = 0.19, p = 0.00; 95% CI 0.07, 0.45). Thus, Hypothesis 6 was supported.
Together, these antecedents explained 64% of the variance of the job insecurity climate. However, our predictors explained different amounts of variance and showed different effect sizes. Our results showed that rumors were the main predictor, with the highest effect and explained variance (27%). After that, in decreasing order, were communication, with an explained variance of 9%, and the percentage of temporary workers in the organization, organizational resilience, and organizational type (7%, 6%, and 4%, respectively). The country was the least important antecedent, with 3% of explained variance.
Discussion
The present study examined the stressor of job insecurity, but unlike the predominant research, not as an individual experience, but as a collective phenomenon at the organizational level; in other words, job insecurity was conceptualized as a collective stressor in organizations (i.e., in terms of job insecurity climate). Specifically, this study portrays how different characteristics of organizations (i.e. type, percentage of temporary employees, organizational resilience), organizational processes (i.e. communication and rumors), and context (i.e., country) can be determinants in the emergence of a job insecurity climate. Indeed, although the job insecurity climate has been noted in the literature, to our knowledge, this study is the first empirical analysis of its antecedents, considering job insecurity as a shared experience.
This study drew from previous research on job insecurity and organizational climate to address the need for a better understanding of antecedents of job insecurity climate. The overall results suggest that organization type, percentage of temporary employees in organizations, organizational resilience, communication, rumors, and country are all related to the job insecurity climate. Thus, the results provided support for the hypotheses. As predicted, a higher job insecurity climate prevailed in private organizations compared to public organizations (hypothesis 1). This was congruent with previous literature on individual job insecurity in public versus private organizations (i.e., Baldwin 1987; Erlinghagen 2008; Mauno & Kinnunen 2002; De Cuyper et al. 2009). Our findings also revealed a positive relationship between the percentage of temporary employees in organizations and their job insecurity climate (hypothesis 2). In line with De Cuyper et als’ (2009) findings, a higher percentage of temporary workers in an organization was associated with a higher job insecurity climate. This result is also congruent with previous research on temporary employment and job insecurity at the individual level for both temporary and permanent workers (Geary 1992; Kinnunen & Nätti 1994; Peiró, Sora & Caballer, 2012; Sverke et al. 2000) according to the segmentation model (Atkinson 1984).
Furthermore, the findings also confirmed the beneficial role of organizational resilience (hypothesis 3). The job insecurity climate was lower in more resilient organizations compared to less resilient ones. Organizations with the ability to respond to unexpected events seemed to provide a more secure environment in terms of job stability for their workers. These results were congruent with previous research that associated organizational resilience with positive organizational outcomes (Bustinza et al. 2019; Hamel & Valikangas 2003; Hollnagel et al. 2006; West, Dyrbye & Sloan 2009), as well as studies that linked organizational resilience to lower job insecurity at an individual level (Prasongthan 2022; Vo-Thanh et al. 2022). Similarly, we hypothesized that communication was negatively associated with the job insecurity climate (hypothesis 4). The results confirmed this hypothesis, showing that adequate communication can act as a way to reduce the overall uncertainty about the continuity of jobs. Moreover, this finding supported the literature on communication in the workplace (Charoensukmongkol & Suthatorn 2022; Robbins 1999; Saeed et al. 2023). Likewise, the findings supported the idea that rumors can promote a job insecurity climate. This also aligns with earlier research on the link between rumors and job insecurity at an individual level (Bordia et al. 2006; DiFonzo & Bordia 2000; DiFonzo, Bordia & Rosnow 1994; Smet et al. 2016), which indicates that rumors are spread to reduce uncertainty about matters of particular relevance for employees. This is the case for overall concern about job loss; thus, an organization with rumors about potential job loss may promote a job insecurity climate within the organization.
Finally, this study provides empirical support for the assumption that the context in which organizations operate is critical in determining the measures they use to ensure job stability or, conversely, to promote job insecurity. Countries with stable and growing economies, strong worker protection laws and institutional and social structures, such as Austria, facilitate job stability and thus, a lower job insecurity climate, compared to Spanish organizations, which reported higher job insecurity climates. These results supported hypothesis 6.
These results suggest that organizational processes and organizational characteristics are critical for the levels of job insecurity climate. In fact, communication, and especially informal communication, such as rumors, represents a key organizational process that can determine the level of overall concern about job loss within an organization. It is, therefore, especially important for organizations to be careful about how they transmit information to their employees in order to avoid rumors and the emergence of a job insecurity climate, which is a collective stressor. Additionally, albeit to a lesser extent, our results indicate that certain organizational characteristics also influence the level of job insecurity climate. Finally, it seems that macro-level factors, such as country, have the weakest – though still significant – effect on the job insecurity climate.
In summary, this study extends the present scope of the job insecurity literature by integrating an organizational perspective into the conceptualization of job insecurity. Furthermore, this study provides preliminary evidence that organizational factors, such as organization type, percentage of temporary employees in the organization, organizational resilience, communication and rumors, may affect the job insecurity climate within an organization, and that rumors and communication play a particularly relevant role. Nevertheless, while the study’s findings are quite supportive of our original arguments, they also suggest that there is still much to be done to better understand the concept of job insecurity climate.
Limitations
A number of limitations need to be kept in mind when interpreting these results. First, although several measures (i.e., the use of native-language instruments and the back-translation method) were adopted to reduce measurement error, there may still be a degree of measurement error. Second, local research teams administered the questionnaires; therefore, convenience sampling could affect our findings. In addition, the samples were non-representative and a convenience sampling method was used. This means that participants were selected based on how easily accessible they were, rather than employing random sampling techniques. Thus, our findings represent the current conditions of specific organizations from particular labor sectors within a defined time frame. Hence, these results need to be interpreted carefully and there are some limitations to their generalizability. Nevertheless, as Sherman (2024) stated, non-representative samples can provide valuable insights and contribute to the understanding of psychological phenomena, especially when resources are limited. In fact, non-representative samples are a practical reality in job insecurity research, often due to the specific research goals, access issues, and resource constraints. To build a more comprehensive understanding of the impact of the job insecurity climate across diverse populations, future research should prioritize representative sampling methods and diverse participant recruitment strategies. It would be interesting to explore these hypotheses in representative samples to provide more empirical support for these results in future research. Third, although we used different sources, we relied on self-reports to assess our variables. This can lead to common variance problems (Avolio, Yammarino & Bass, 1991; Brown & Benson 2005). Fourth, data were collected in 2011 to study organizations in the context of global crises, such as the economic crisis of 2009. During an economic crisis, many organizations must adopt specific measures, such as downsizing, to survive and be competitive in this restrictive context. Hence, job insecurity emerges as a general concern among employees. Studying the job insecurity climate in this specific period can facilitate a better understanding of this phenomenon. In fact, if we examine the characteristics of the current labor market compared to that in 2011, we see that the European labor market has made substantial progress since then, resulting in greater robustness and stability. The EU-27 unemployment rate has decreased from 9.9% in 2011 to 6.0% in 2023, and temporary employment has also decreased from 14.5% in 2011 to 13.8% in 2022. Accordingly, our results reflect the job insecurity climate at a specific time with specific characteristics, and it is possible that they may not be generalizable to contemporary society. However, our findings contribute to the theoretical understanding of job insecurity climate and provide a valuable historical perspective. Nevertheless, further research is necessary to explore the job insecurity climate in other economic contexts. Fifth, this study has a cross-sectional design, which limits the ability to draw conclusions on causal relationships. Finally, this study adopted an approach at the organizational level by aggregating and analyzing all variables at this level. This perspective does not allow for the analysis of cross-level effects; in other words, the simultaneous effect at multiple levels (e.g., individual and organizational). This might mask important individual-level predictors or interactions that affect the job insecurity climate differently within organizations. Future research could adopt a multilevel perspective to explore how individual variables interact with the organizational context to shape the job insecurity climate.
Theoretical and practical implications. These results provide theoretical implications by (1) providing additional evidence about the emergence of job insecurity as a collective phenomenon within organizations (job insecurity climate), and by (2) examining the factors that are critical in determining the levels of job insecurity climate within an organization. Regarding the practical implications, this study is valuable because it enables managers and HR professionals to design actions to prevent and address job insecurity within their organizations by improving some critical organizational characteristics: the percentage of temporary employees, organizational resilience, communication processes, and rumors. For example, reducing the dependence on temporary contracts and clarifying pathways toward permanent employment may promote individual and collective job security. Also, establishing transparent and bidirectional communication channels is essential to effectively disseminate accurate information and minimize the spread of rumors about the possibility of job loss within organizations.
Future research. This study focused on specific organizational factors and countries to explain job insecurity climate. However, we also acknowledge that there can be multiple potential antecedents to job insecurity climate that we did not assess. For example, we used the country to capture the contextual characteristics in which organizations are operating. Future research is necessary to clarify the relationship between these contextual characteristics (i.e. economic growth, welfare state typologies, unemployment, etc.) and the job insecurity climate more specifically. In addition, future studies could also explore a broader range of organizational-level antecedents, such as leadership styles. Employee-oriented leadership may reduce the job insecurity climate by fostering trust and support. Another characteristic could be the use of downsizing measures. The literature has established that downsizing is associated with higher levels of individual job insecurity. It seems plausible to propose that given that it is an overall measure adopted in the organization, the experience of job insecurity could be shared among employees as well. In addition, this research focused on the content of the job insecurity climate, overlooking its strength. Future research could explore the antecedents of both the content and strength of job insecurity climate and clarify whether they are different. Our study provides a baseline for understanding job insecurity climate and its organizational antecedents by focusing on organizational characteristics as a direct effect. However, it could be beneficial to identify the underlying mechanisms that could explain the association between organizational factors and job insecurity climate. Future research could examine the possible intervening variables that provide better knowledge about these relationships. Finally, research on job insecurity at the individual level has demonstrated that there are reciprocal relationships between job insecurity and its correlates (e.g., Jiang & Lavaysse 2018; Nawrocka et al. 2021). Regarding job insecurity climate, the only study we are aware of is by Låstad et al. (2016), which showed the reciprocal relationship between job insecurity climate and individual job insecurity. According to this research, it is possible to suggest that the relationship between some of our antecedents, especially organizational resilience, communication and rumors, could be reciprocal. Further research with a longitudinal design is needed to clarify this issue.
In conclusion, this study evidenced that job insecurity is not only an individual experience but can also emerge as a collective phenomenon within organizations. In addition, different factors were identified as critical for determining the level of the job insecurity climate within an organization: type of organization, percentage of temporary employees, organizational resilience, communication, rumors, and country. Among these, rumors were especially noteworthy due to their robust and strong effect. These findings not only offer valuable insights for academics but they may also be relevant for practitioners. Understanding these dynamics can facilitate the development of effective organizational interventions aimed at fostering healthier organizational climates and mitigating collective job insecurity.
Competing Interests
The authors have no competing interests to declare.