Comparing working conditions and job satisfaction in hospitality workers across Europe

Highlights

• •The study explores the influence of the institutional context on job satisfaction.
• •The research develops a multilevel analysis that considers institutional and organizational factors.
• •Data of 22792 employees of 16 European countries from the Sixth European Working Conditions Survey are part of the sample.
• •A novel classification of countries according to their working conditions is provided.

Abstract

Job satisfaction is important in the tourism sector since workers’ satisfaction is key to providing high-quality service, which is very important in determining organizational success. The working conditions that influence job satisfaction depend to a large extent on the institutional context, which shows similarities in some European countries. This research aims to compare working conditions and job satisfaction among European country blocks that have similar institutional characteristics. Unlike previous studies, this research adopts a comprehensive approach by considering institutional and organizational factors in the analysis of employees’ perceptions of job satisfaction. The sample is made up of 1633 workers in 16 European countries. The results demonstrate the existence of three different models of working conditions in Europe leading to differing levels of job satisfaction in tourism. These models do not correspond to the clusters identified by the previous literature, which adopts an institutional perspective.

1. Introduction

Satisfaction at work is one of the most studied topics in the management literature (Dixit and Dean, 2018; Jung and Takeuchi, 2018; Lee and Chelladurai, 2018). Job satisfaction not only affects the productivity and performance of workers, it also influences how a company’s goals are achieved in terms of improving customer satisfaction, perceived service quality, customer loyalty and satisfaction, and brand image (O’Donoghue and Tsui, 2013). This is especially relevant in the service industry since an adequate quality of service involves employee attitudes and behaviors that affect customers’ experiences and expectations (Oliver, 1980). Our research focuses on the tourism sector due to the strong position this sector has in the European economy. According to the United Nations World Tourism Organization, the tourism sector comprises the third largest economic activity in the European Union (EU), accounting for about 10 % of its gross domestic product and ranking as the fourth sector in terms of exports. Tourism also contributes employment in the EU equivalent to 9.7 % of total employment (World Tourism Organization, 2018). Despite its importance for the European economy, the tourism sector is characterized by underpaid jobs and high work-related stress (Jovanović et al., 2019; Lillo-Bañuls et al., 2018). This is due to the characteristics of jobs in this sector, where limited career opportunities and broad work schedules exist (Hofmann and Stokburger-Sauer, 2017; Stamolampros et al., 2019).

Job satisfaction is the result of different factors, among which working conditions play an important role. As shown in the literature, working conditions such as salary, promotional possibilities, job security, and the working climate highly affect job satisfaction (Dalkrani and Dimitriadis, 2018). At the same time, working conditions are determined by different factors, among which the institutional context becomes especially relevant (Boon et al., 2009). The literature defines the institutional environment as the set of directives, rules, laws, norms, and legal standards that determine the normative structure for economic and social development (Acemoglu and Johnson, 2005; DiMaggio and Powell, 1983; Scott, 1987). The institutional context is determined by different factors such as economic conditions, unemployment rate, the national level of income inequality, and the degree of unionization (Pichler and Wallace, 2008). However, the influence that the institutional context has on working conditions among countries classified as similar in terms of their institutional setting can be heterogeneous because, although the institutional context establishes the framework in which working conditions are developed, these conditions highly depend on employment practices implemented by organizations.

Job satisfaction is a concept that has been measured both nationally and internationally (Lee and Chelladurai, 2018) in different sectors such as in banking and the public or hospitality sector (Ariza-Montes et al., 2018; Kong et al., 2018). Most of these studies were conducted without taking into account the institutional context. Economic conditions, unemployment rate, and national level of inequality of a national territory, among other institutional factors, generate similar working conditions among countries in terms of salaries, working hours, job security, and flexibility (Posada-Kubissa, 2018; Tangian, 2008). Working conditions are particularly context‐sensitive due to their strong linkage to the industrial relations system of a country, unemployment rate, etc. (Van Dierendonck et al., 2016).

Despite the existence of a supranational government in the EU, the institutional context differs across countries, and therefore working conditions and employee satisfaction are also different across Europe. Previous studies have classified countries according to their institutional context and identify different models of human resource management in Europe (e.g., Brewster and Tregaskis, 2003; Ignjatović and Svetlik, 2003; Nikandrou et al., 2005). It is interesting to complement these studies that present an institutional focus with a perspective centered on organizational practices and employees´ perceptions. For employment practices to create value for companies and society, they must generate job satisfaction. Due to the importance of job satisfaction at individual, organizational, and societal levels, including employees´ perception of their job satisfaction, the analysis becomes crucial. A deep understanding of the differences in job satisfaction across Europe could set the basis for a deeper discussion and formulation of novel hypotheses regarding the influence of institutional factors on working conditions. This understanding could lead companies and policy-makers to propose policies for improving working conditions in order to enhance job satisfaction and social welfare.

Although some studies that compare job satisfaction across European countries can be found in the literature (e.g., Eskildsen et al., 2004; Millán et al., 2013; Pichler and Wallace, 2008), comparisons are made across national territories without considering the homogeneity that may exist among European countries. According to the literature, these countries can be grouped by blocks according to the similarities in their approach to the welfare state—which impacts, among its main facets, working conditions. The welfare state model of each country is determined, among other aspects, by public policies, labor regulation, and organizational practices—fundamentally, human resources management practices. Hence, the literature establishes blocks of countries based on their similarities in their institutional setting and their prevailing organizational human resources management models (e.g., Albareda et al., 2007; Brookes and Barfoot, 2005; Filella, 1991; Ronen and Shenkar, 1985; Tangian, 2008). The underlying premise is that there is some convergence toward homogeneity of these characteristics of countries within the same cluster and differences with respect to the rest of the blocks. Studies that analyze whether this convergence leads to homogeneity in workers´ perception of labor conditions and job satisfaction across Europe are rare.

This study tries to contribute to this end by exploring working condition models in Europe from an organizational perspective and considering workers´ perceptions. This might allow identification of possible deviations between the institutionally established regulations at the national or supranational level and the patterns of interaction of the workers and organizations in the labor market. This can help us understand which models lead to higher levels of job satisfaction and whether there is convergence in this aspect in the European context. The research seeks: (i) to analyze the different models of working conditions—what likely leads to differences in perceived job satisfaction—that exist in Europe; and (ii) to explore whether these models differ among the clusters of countries based on institutional characteristics identified in the previous literature. From these objectives, the following research question is derived: Does the clustering of European countries according to institutional characteristics correctly reflect the differences in labor conditions and subsequently job satisfaction across Europe?

This article is divided into six sections. First, a review of the relevant literature is presented in the second section. Next, the methodology of the research and the results are explained in the third and fourth sections. Finally, a discussion of the results and the conclusions, which includes the limitations and suggestions for future research, are detailed in the fifth and sixth sections.

2. Literature review

2.1. Job satisfaction and working conditions in the tourism sector

Job satisfaction is an essential aspect for firms to gain a competitive advantage in all sectors, given the central role that employees play in business success (Kramar, 2014). However, despite the importance of job satisfaction, there is no general agreement regarding its definition. Different authors have contributed to its clarification. Among the most-cited definitions is the one given by Spector (1997), who emphasizes that job satisfaction refers to the way employees feel about their job and depends on different factors. Mahdieh and Sotoudehnama (2018) affirm that job satisfaction depends on factors such as personal, organizational, managerial, academic, professional, and economic variables. Goetz et al. (2016) underline four factors as determinants of job satisfaction: professional development, interpersonal relations, economic expectations, and working conditions.

There are principally two methodologies for assessing job satisfaction: the integral measurement of a single factor and the comprehensive multidimensional measurement. The difference between the two methods lies in the fact that while the former relies on a single item to measure job satisfaction, the latter employs several factors. Most research on job satisfaction at the national level adopts a multidimensional measurement approach. For instance, the descriptive work index (JDI) developed by Locke et al. (1964) includes different dimensions of the job such as promotion, payment, and relationships with managers and colleagues. Spector (1997) created a Job Satisfaction Survey (JSS) that contains nine dimensions: salary, promotions, additional benefits, incentives, superiors, colleagues, operating environment, intrinsic work characteristics, and communication. Parent-Thirion et al. (2016) developed their Job Quality Index (JQI) from seven variables (earnings, prospect, social environment, physical environment, work intensity, skills and discretion, and work time quality) that are related to the multidimensional nature of work. The JQI has been considered for the present investigation because it is comprehensive in coverage, transparent in method, and widely employed in the research on job satisfaction and the quality of work in the European context (e.g., Erro-Garcés and Ferreira, 2019; Punzo et al., 2018; Soriano et al., 2018). It is the basis for the development of the sixth European Working Conditions Survey (EWCS) which, according to Grimshaw et al. (2017), yields solid and reliable information. In 2000, the EU launched the European Employment Strategy with the aim of creating more (quantity) and better (quality) jobs (Ariza-Montes et al., 2019). EWCS asks workers about the intrinsic characteristics of their jobs: salary, hours, participation, organization, and security, among others. The EWCS has been used in previous studies in which the impact of working conditions on satisfaction is analyzed, but using different perspectives such as new technologies (Castellacci and Viñas-Bardolet, 2019), gender issues (Brinck et al., 2019; Gómez-Baya et al., 2018), and workers’ age (Berde and Rigó, 2020; Okay-Somerville et al., 2019).

The tourism sector is characterized by high levels of seasonality, which leads to labor practices that do not favor workers’ commitment and permanence in the company in the long term (Hofmann and Stokburger-Sauer, 2017). The characteristics of the job positions in the tourism sector are related to higher levels of job dissatisfaction compared to other industries, which explains why more than half of the workers in the tourism sector are dissatisfied and consider moving to other sectors (Stamolampros et al., 2019). Factors explaining the low levels of job satisfaction observed in the tourism sector are related to characteristics of job positions and to the lack of professionalization of the human resources management in this industry (Jovanović et al., 2019; Lillo-Bañuls et al., 2018; Zopiatis et al., 2014). On the one hand, the characteristics that make this sector present low levels of job satisfaction compared to other sectors are related to low salaries (earnings), long working hours (work intensity), low job security, and the scarcity of promotional possibilities (prospects) (Zopiatis et al., 2014). The low work time quality of the jobs in the tourism sector is associated with the continuous relationship with customers, shift work, unsocial hours, and night work (Lillo-Bañuls et al., 2018). This, together with the scarcity of occupational health and safety practices that favor an adequate physical environment, make employees working in this sector experience difficulties maintaining a work-life balance and a healthy lifestyle that would prevent stress and not lead to low levels of job satisfaction (Hofmann and Stokburger-Sauer, 2017). This stress is increased by the lack of perceived organizational support and autonomy that characterize jobs in the tourism industry (Loi et al., 2014; Tongchaiprasit and Ariyabuddhiphongs, 2016). The low levels of employee recognition, centralization in decision-making, and presentism that characterize this industry are associated with a lack of professionalization of human resources in the tourism industry (Nickson, 2013). The degree to which employees perceive social support from their superiors (the quality of the social environment at work) and are provided with autonomy to perform their job (skills and discretion) highly determine employees’ level of satisfaction and work engagement since the social support of managers and supervisors influence workers´ perception of justice at the workplace (Jovanović et al., 2019).

To obtain a comprehensive view of job satisfaction and its antecedents in the tourism sector, different dimensions must be considered. This study combines different factors that determine the quality of work (earnings, prospect, social environment, physical environment, work intensity, skills and discretion, and work time quality) to provide a holistic view of working condition that allows the comparison of the quality of work and the level of job satisfaction across Europe by relying on the employee’s own perspective.

2.2. Influence of the institutional context in job satisfaction

The EU´s regulations favor workers’ mobility within Europe. Labor mobility is the result of different levels of national unemployment rates, salary level, flexibility, etc. (Fahri and Werning, 2014). Taking into account that the quality of employment varies across European countries, factors that strongly explain workers ´ mobility and differences in job satisfaction depending on the country can be observed, as indicated in the literature (e.g., Leineweber et al., 2016; Salpigktidis et al., 2016; Thite et al., 2012). These differences can be explained by the distinct institutional settings of each territory (Salvatori, 2010). As derived from the premises of institutional theory, coercive pressures—especially national regulations—highly determine human resource management practices, so they might lead to differences in working conditions across countries (Western, 1998). The different labor legislations across European territories, despite European countries sharing a supranational government, influence working conditions and job satisfaction (Brewster and Hegewisch, 2017). According to institutional theory, in addition to the coercive pressures exerted by legislation in a country, there are normative pressures, which are related to the appropriate and desirable norms of behavior for both organizations and individuals that predominate in a country (Acemoglu and Johnson, 2005; DiMaggio and Powell, 1983; Scott, 1987). These pressures also vary across territories and can be determinant in working conditions. Countries that present similar institutional contexts—that show similar coercive and normative pressures—might present differences in terms of employee job satisfaction. This could be the case in countries such as Denmark and Norway, which present both institutional and cultural similarities but significantly differ in their working conditions (Bech et al., 2017).

Reviewing the literature, it can be observed that previous research has made efforts to identify blocks of European countries according to their institutional context (e.g., Albareda et al., 2007; Brookes and Barfoot, 2005; Filella, 1991; Ronen and Shenkar, 1985; Tangian, 2008). One of the most commonly used classifications identifies four clusters of countries in Europe: Anglo-Saxon (Ireland and the United Kingdom), Central European (Austria, Belgium, Germany, the Netherlands, and Switzerland), Latin (France, Greece, Italy, Portugal and Spain) and Nordic (Denmark, Finland, Norway and Sweden) (Filella, 1991; Ronen and Shenkar, 1985).

Numerous aspects of institutional context determine working conditions. Pichler and Wallace (2008) emphasize the key role played by four institutional factors in working conditions: economic conditions, unemployment rate, the national level of inequality, and the degree of unionization. Economic conditions of a territory highly impact the labor market in terms of job rewards in both extrinsic (average wage level, working hours, etc.) and intrinsic terms (meaningful, high-skilled jobs, etc.). The national unemployment rate and the national level of inequality also influence working conditions and job satisfaction. High levels of unemployment hinder job mobility regardless of a workers’ level of satisfaction. Employees, even those who are dissatisfied, will remain in their jobs because of the lack of opportunities in the labor market. The scarcity of job opportunities and the excess of job demand might lead employers to offer poorer conditions in terms of salary, working hours, etc. Socio-economic inequality is also a determinant of job dissatisfaction if employees perceive that similar jobs lead to great differences in economic outcomes. The degree of unionization in a country seems to be highly determinant of the average wage level and other conditions of work that influence the welfare of employees. In highly unionized countries, employees are more likely to find better jobs in terms of salary, working hours, etc. Accordingly, working conditions are generally better in countries that present a solid economic situation, a low unemployment rate, and a high level of unionization. This is the case for companies in the Nordic cluster, which have good working conditions in comparison with the rest of European companies (Eskildsen et al., 2004). This can be explained by the high level of trade union intervention in those countries, where labor reforms encourage workers’ representatives to negotiate working conditions with trade unions.

As indicated in the literature, another institutional characteristic that determines working conditions is the country level of regulation (Gialis et al., 2017; Keune and Jepsen, 2007). The level of regulation is closely related to the level of flexibility in the labor market and to the degree of job security (Posada-Kubissa, 2018). Labor flexibility is negatively associated with job satisfaction and employees´ physical and psychological health since flexibility is associated with low levels of job security (Carr and Chung, 2014; Probst et al., 2017). Flexibilization comes from deregulation; job security pursues the maintenance of social advantages through a compensatory system. Both depend on the country and are not only affected by economic conditions, but by collective agreements, and by the agents involved: governments, employers, and trade unions (Tangian, 2007). In this line, Sapir et al. (2004) identified four different social systems within Europe according to the level of flexibility of each country. Gil-Alana et al. (2019) affirm that a robust social security system is associated with low levels of inequality.

From the aforementioned two premises are derived: (i) that the institutional context strongly influences working conditions and that these become a determinant factor in job satisfaction (Williams and Hall, 2000); and (ii) that since institutional pressures are similar in each country block—Anglo-Saxon, Central European, Latin, and Nordic—similar working conditions within each cluster (intra-group similarities) and differences across clusters are expected (inter-groups differences). This is because, among other aspects, government regulations determine an organization’s freedom of action regarding employees’ minimum wages, training and development investments, working hours, etc. (Vaiman and Brewster, 2015).

Although the influence of the institutional context on working conditions is expected, companies’ freedom of action within the framework of labor regulations is also expected to determine working conditions. In this way, workers in the tourism sector of countries with similar institutional settings could present discrepancies in their working conditions and, subsequently, in their job satisfaction. Providing evidence about this would justify the need to group countries according to their working conditions model, a categorization that would more accurately show the reality of the labor market from an employee's perspective. In order to address the research objectives, the methodology used to develop the empirical analysis is presented below.

3. Sample and method of research

3.1. Sample

To investigate differences in job satisfaction and in the quality of work among countries that show significant institutional differences, we have focused on the tourism sector due to the relevant role it plays in the European economy. The data used for the research were extracted from the sixth EWCS (the most recent available). This survey contains data on 43,850 working individuals 15 years old or older residing in private homes in one of the 33 European countries studied (28 countries of the EU; Albania; the former Yugoslav Republic countries of Macedonia, Montenegro, and Serbia; and Turkey). This survey was developed by the European Foundation for the Improvement of Living and Working Conditions (2020) (dependent on the European Commission) to obtain information on the quality of work and employment in Europe.

To perform the analysis, countries that present significant institutional and organizational differences were selected (Filella, 1991; Ronen and Shenkar, 1985). The sample includes the following countries and country clusters: the United Kingdom and Ireland (Anglo-Saxon); Austria, Belgium, Germany, the Netherlands and Switzerland (Central European); France, Greece, Italy, Portugal, and Spain (Latin); and Denmark, Finland, Norway and Sweden (Nordic). The sample used in this study is formed of 1633 employees of 16 European countries that work in the tourism sector. Table 1 shows the number of observations for each country cluster.

Table 1.

Tourism sector sample (country clusters).

Country cluster	Sample size
Anglo-Saxon	181
Central European	503
Latin	769
Nordic	180
Total	1633

To select workers from the tourism sector, the Statistical Classification of Economic Activities in the European Community, NACE1 codes were used. According to Eurostat, the following codes were included as part of the tourism sector: 491 (Passenger rail transport and interurban); 493 (Other passenger land transport); 501 (Sea and coastal passenger water transport); 503 (Inland passenger water transport); 511 (Passenger air transport); 551 (Hotels and similar accommodation); 552 (Holiday and other short-stay accommodation); 553 (Campgrounds recreational vehicle parks and trailer parks); 561 (Restaurants and mobile food service activities); 563 (Beverage serving activities); 772 (Rental and leasing of personal and household goods); 791 (Travel agency and tour operator activities); and 799 (Other reservation service and related activities). Filtering by these criteria, 1633 employees (7.2 % of 43,850) made up the sample.

3.2. Variables

We based our research on the sixth edition of the EWCS, which includes the dimensions of the European JQI developed by Parent-Thirion et al. (2016). This index is formed of seven dimensions that determine working conditions: earnings, prospects, social environment, physical environment, work intensity, skills and discretion, and work time quality.2 All the constructs used in the analysis except salary (expressed in euros) and job satisfaction (expressed on a four-point Likert scale) are numerical variables expressed on a scale of values between 0 and 100. According to Parent-Thirion et al. (2016), the constructs were defined as follows:

Earnings: The importance of earnings as a motivational factor has been widely studied in the literature (Suzuki et al., 2018). This construct is defined as the net hourly earnings of workers.

Prospects: This refers to the job characteristics that contribute to a person’s material and psychological needs, encompassing the need for income and for employment continuity. De Witte et al. (2016) point to these factors as determinants of job satisfaction.

Skill and discretion: This dimension refer to the skills required for the job and the level of job autonomy. Both are pointed to in the literature as relevant factors influencing job satisfaction since they enhance job identification and commitment (Fregin et al., 2018; Mateos-Romero and del Mar Salinas-Jiménez, 2018).

Social environment: This dimension measures the social support perceived by employees (good social relations with line managers and fellow workers) and the absence of abuse in the company, which becomes especially important for workers’ welfare as it moderates the negative impact of stressors (Wisse et al., 2018). This construct includes two constructs: adverse social behavior and social support.

Physical environment: This dimension refers to environmental hazards and to factors related to posture-related risks, which become relevant factors in the health of employees, a fundamental aspect of job hygiene and satisfaction (Devonish, 2018; Koh et al., 2017).

Work intensity: This dimension refers to the intensity of work demands. High work intensity is associated with a risk of suffering high levels of occupational stress, which in turn is associated with low levels of job satisfaction (Iranmanesh et al., 2017; Rushton et al., 2015).

Work time quality: This dimension refers to the organization and length of working time. The number of working hours, shift work, night work, etc., are determinant for the achievement of a good work/life balance, subsequently playing a significant role in job satisfaction (Eagan et al., 2015; Roy, 2017).

Job satisfaction: The level of satisfaction is a variable included in the Sixth EWCS survey and is measured as a four-point Likert scale. The question is: “In general, are you very satisfied, satisfied, not very satisfied or not at all satisfied with your working conditions?”.

All the items used for the construction of the variables are included in the sixth EWCS and are shown in the appendix, together with the results of the reliability tests obtained with the Cronbach alpha coefficient for the tourism industry.

3.3. Method of analysis

The main objective of the empirical analysis is to determine whether the classification of countries based on the institutional context adequately reflects the different models of working conditions—and subsequently differing levels of job satisfaction—existing in Europe in the tourism sector, and if not, to propose a more appropriate classification of countries. To do this, based on the classifications of Filella (1991) and Ronen and Shenkar (1985), a comparison of working conditions among countries of the same clusters (intra-group comparison) is made. The existence of a high heterogeneity among countries of the same block would indicate an inappropriate grouping of countries located within the same institutional block. This analysis will be completed with an inter-group comparison, in which a high homogeneity in the working conditions of countries of different blocks would indicate a reduced discriminatory capacity among the blocks. Therefore, a high intra-group heterogeneity and a reduced inter-group heterogeneity would allow us to conclude that the classification made by previous studies does not correctly classify countries according to the labor conditions perceived by workers. Next, through a two-step cluster analysis, a new classification is proposed that improves intra-group homogeneity and inter-group heterogeneity. The suitability of this new group of countries will be evaluated using the methods previously described.

The normality of these variables was previously checked for the selection of the method of analysis. To address the research objectives, both inter-group and intra-group differences have been analyzed for both job satisfaction and working conditions. First, the analysis of inter-group differences—among country blocks—has been performed using the Mann-Whitney test. This technique allowed a comparison of the level of job satisfaction among country clusters (Anglo-Saxon, Central European, Latin and Nordic). Second, the existence of significant intra-group differences among countries within the same cluster in the level of job satisfaction have been studied using the Mann-Whitney and Kruskal-Wallis tests due to the ordinal nature of this variable. As the Mann-Whitney test can only be used to make comparisons between two groups, it has been employed to test the intra-group differences in the level of satisfaction within the Anglo-Saxon cluster (between Ireland and the United Kingdom). Since the Kruskal-Wallis test allows comparing more than two groups, it was used to analyze the existence of intra-group differences for the Central European, Latin, and Nordic clusters.

Working conditions have also been compared among country blocks (inter-groups) and among countries within the same block (intra-groups). First, the analysis of inter-group differences in working conditions has been performed using the t-Student test. This technique allowed the comparison of the working conditions among all the country blocks. Second, as working conditions (earnings, prospects, social environment, physical environment, work intensity, skills and discretion, and work time quality) are numeric variables and normally distributed, t-Student and analysis of variance (ANOVA) have been used to analyze the intra-group differences. As the t-Student test can only be used to compare two groups, it was employed to analyze the intra-group differences of working conditions within the Anglo-Saxon cluster. As ANOVA allows comparisons among more than two groups, it was used to assess the existence of intra-group differences among the Central European, Latin, and Nordic blocks.

The effect sizes have been estimated with the statistic proposed by Rosenthal (1994) for the Mann-Whitney contrasts (0.1, 0.3, and 0.5 are used to indicate small, medium, and large effect sizes); Cohen’s d statistic for t-Student contrast (0.2, 0.5, and 0.8 are used to indicate small, medium, and large effect sizes), and η² statistic for the ANOVA test (0.01, 0.06, and 0.14 are used to indicate small, medium, and large effect sizes) proposed by Cohen (1977). A ${\epsilon }^2$ statistic is used for the Kruskal-Wallis test (0.01, 0.08, and 0.26 are used to indicate small, medium, and large effect sizes) (Tomczak and Tomczak, 2014).

The existence of significant intra-group differences and limited differences among blocks of countries that present different institutional settings justifies the need for a new classification of European countries. To create this new grouping, a two-step cluster analysis has been developed. To confirm the validity of the proposed clusters, the intra-group and inter-group differences in the level of job satisfaction and in working conditions have been analyzed using the same statistical techniques previously explained.

4. Results

The descriptive analysis of the data shows that the average age of employees of the sample is heterogeneous, standing at just over 40 years, with a standard deviation of 12.92 years. The male gender is slightly predominant; they represent 56.2 %, compared to 43.8 % of women, which contrasts with the existing proportion in this sector at the European level, where these proportions are inverse. Secondary education is the predominant level of education among workers in the sample (74.1 %), followed by university studies (19.1 %), and primary education (6.8 %). The most represented sub-sectors in the sample are "beverage serving activities," which represent the majority group (54.8 %), "passenger rail transport and interurban" and "other passenger land transport" (21.2 %), and accommodation ("hotels and similar accommodation," "holiday and other short-stay accommodation,” and “campgrounds recreational vehicle parks and trailer parks”) (15.1 %). Following the International Standard Classification of Occupations (ISCO-08) based on OECD (2012), 71.8 % of the workers in the sample are “white collar” employees, of which less than a quarter are highly qualified. Within the “blue collar” employees—who represent 28.2 % of the total sample—only 7.5 % are considered highly qualified.

Presuming that the institutional environment is a factor that could significantly affect the degree of satisfaction of workers, in particular those who work in the tourism sector, we have explored the levels of job satisfaction across country blocks that present institutional differences. Using the Mann-Whitney test, we analyzed the differences among working conditions in country blocks with different institutional contexts. The results show that there are mainly significant differences in the level of satisfaction in the Latin countries with respect to the rest of the blocks, while the differences among the rest of the blocks are not significant. In addition, the effect size is very small, even in the case where the differences are significant (Table 2 ). Hence, there is a high homogeneity in job satisfaction across country blocks that present different institutional settings.

Table 2.

Inter-groups differences in job satisfaction. Mann-Whitney test. P-value (Effect size).

Variable	Ang -Nor	Ang -Cen	Ang -Lat	Nor -Cen	Nor -Lat	Cen-Lat
Job satisfaction	0.083 (0.092)	0.501 (0.026)	0.000 (0.155)	0.121 (0.060)	0.006 (0.089)	0.000 (0.175)

When analyzing the intra-block differences, within the Nordic cluster, Denmark and Finland do not present any unsatisfied employees. About 90 % of employees present high and medium-high levels of satisfaction in Austria and Switzerland (within the Central European cluster), the latter not presenting any unsatisfied employees (see Table 3 ).

Table 3.

Intra-group differences in job satisfaction. Kruskal-Wallis and Mann-Whitney tests. P-value (Effect size).

Country	Very satisfied	Satisfied	Not very satisfied	Not at all satisfied	Intra-group comp p-value (E.S.)
United Kingdom	34 %	49.5 %	13.4 %	3.1 %	0.416a (0.061)
Ireland	40.2 %	45.1 %	12.2 %	2.4 %
Denmark	21.9 %	71.9 %	6.3 %	0 %	0.003 (0.079)
Finland	28.6 %	50 %	21.4 %	0 %
Norway	38.5 %	55.8 %	3.8 %	1.9 %
Sweden	13 %	611 %	16.7 %	9.3 %
Austria	44.7 %	47.4 %	3.9 %	3.9 %	0.001 (0.038)
Belgium	22.7 %	61.3 %	10 %	6 %
Germany	23.8 %	63.5 %	11.1 %	1.6 %
Netherlands	34.5 %	52.7 %	9.1 %	3.6 %
Switzerland	40.7 %	50.5 %	8.8 %	0 %
France	21.1 %	61.1 %	15.8 %	2.1 %	0.277 (0.007)
Greece	13 %	64.9 %	19.1 %	3.1 %
Italy	18.4 %	62.2 %	14.3 %	5.1 %
Portugal	15.5 %	72.6 %	10.7 %	1.2 %
Spain	19.5 %	53 %	21.8 %	5.7 %

^aMann-Whitney test (two countries).

The analysis of intra-group differences shows that these differences are significant; therefore, a lack of homogeneity in job satisfaction among countries in the same block is observed, mainly in the Central European and Nordic blocks, in which the effect size is medium. Accordingly, differences in the degree of job satisfaction among countries within the same block are found, indicating high intra-group heterogeneity.

The differences in working conditions among the blocks of countries identified in the literature—based on their institutional characteristics—were also studied. The results show that the Latin cluster presents significant differences with respect to the rest of the blocks in all the analyzed variables (except the social environment variable), with some effects of medium size. The results show the absence of significant differences between the Anglo-Saxon cluster and the Central European block in all the variables studied. The same is observed when comparing the former with the Nordic group, except in the labor expectations and the physical environment variables, although with a small effect. The differences between the Nordic and the Central European blocks are reduced since, in addition to finding differences in the previous variables, significant differences are also observed in the skills needed to develop the work, although with a small effect (see Table 4 ). Hence, there is a high homogeneity in the working conditions across country blocks that present different institutional settings.

Table 4.

Inter-group differences in working conditions. T-Student-test. P-value (Effect size).

Variables	Ang -Nor	Ang -Cen	Ang -Lat	Nor -Cen	Nor -Lat	Cen-Lat
Earnings	0.062 (0.211)	0.192 (0.124)	0.022 (0.279)	0.328 (0.090)	0.000 (0.591)	0.000 (0.425)
Prospects	0.015 (0.256)	0.327 (0.089)	0.000 (0.373)	0.042 (0.188)	0.000 (0.630)	0.000 (0.475)
Social environment	0.737 (0.037)	0.436 (0.071)	0.370 (0.084)	0.694 (0.035)	0.142 (0.124)	0.009 (0.158)
Physical environment	0.001 (0.339)	0.651 (0.039)	0.001 (0.276)	0.000 (0.393)	0.649 (0.038)	0.000 (0.323)
Intensity	0.129 (0.160)	0.978 (0.003)	0.000 (0.338)	0.056 (0.166)	0.012 (0.193)	0.000 (0.350)
Skills and discretions	0.193 (0.137)	0.069 (0.158)	0.000 (0.303)	0.001 (0.296)	0.000 (0.439)	0.010 (0.148)
Work time quality	0.609 (0.054)	0.168 (0.120)	0.020 (0.192)	0.437 (0.068)	0.003 (0.250)	0.000 (0.316)

Ang: Anglo-Saxon; Cen: Central European; Lat: Latin; Nor: Nordic.

Comparing the working conditions of the countries within each cluster, Ireland and the United Kingdom (Anglo-Saxon block) show a great homogeneity in all variables except salaries—workers in the tourism sector in the United Kingdom receive higher salaries than in Ireland. However, differences among countries of the same block are significant if we analyze the rest of the blocks, as can be extracted from the results of the intra-group ANOVA test (see Table 5 ). Among the Nordic countries, significant differences are observed in the prospects, physical environment, work intensity, and skills and discretion variables. The differences found among the countries of Central Europe are also significant. A high disparity in wages across countries within this block can be observed, motivated by the high average salary in Switzerland, followed by the significant differences in job prospects, in the social environment, and in the skills required for the jobs. Hence, the results indicate the existence of a high degree of heterogeneity in the working conditions of countries within the same block.

Table 5.

Intra-group differences in working conditions. ANOVA and T-Student tests. P-value (Effect size).

Variables	Anglo-Saxon	Nordic	Central European	Latin
Earnings	0.0431 (0.336)	0.355 (0.020)	0.000 (0.182)	0.000 (0.114)
Prospects	0.1481 (0.217)	0.044 (0.045)	0.009 (0.027)	0.000 (0.087)
Social environment	0.0621 (0.296)	0.750 (0.007)	0.029 (0.023)	0.000 (0.041)
Physical environment	0.0681 (0.274)	0.033 (0.048)	0.731 (0.004)	0.000 (0.093)
Intensity	0.1371 (0.223)	0.019 (0.055)	0.634 (0.005)	0.000 (0.082)
Skills and discretion	0.9191 (0.015)	0.008 (0.065)	0.002 (0.065)	0.000 (0.032)
Work time quality	0.4121 (0.123)	0.094 (0.036)	0.575 (0.006)	0.000 (0.042)

¹p-value of t-Student test (two countries).

Based on the previous results which show differences in working conditions among the countries of the same block and scarce differences among blocks established according to their institutional characteristics (with the exception of the Latin cluster), we propose the creation of a classification of countries according to the similarity in their working conditions in the tourism sector, specifically from the seven JQI dimensions (earnings, prospects, social environment, physical environment, work intensity, skill and discretion, and work time quality). To create this new clustering, a two-step cluster analysis was performed (see Table 6 ).

Table 6.

Proposed country blocks according to their working conditions in the tourism sector.

Group	Countries
Group 1	Austria – Belgium – Germany – Italy – Ireland– Netherlands – Norway – Portugal – Switzerland – United Kingdom
Group 2	Denmark – Finland – France – Sweden
Group 3	Greece – Spain
Variables	Group 1	Group 2	Group 3
Work intensity
Physical environment
Work time quality
Prospects
Skills and discretion
Social environment
Earnings

The results of the cluster analysis show that, on the one hand, there are countries such as Greece and Spain (group 3) that show worse working conditions and, consequently, lower levels of job satisfaction in comparison with the rest of the countries. At the other extreme are Denmark, Finland, France, and Sweden (group 2), which present the most advantageous working conditions and the highest degree of job satisfaction. Finally, an intermediate group (group 1) including the rest of the countries can be found (Austria, Belgium, Germany, Italy, Ireland, the Netherlands, Norway, Portugal, Switzerland, and the United Kingdom).

To confirm the validity of these results (average silhouette value is greater than 0.5), the working conditions of the groups created and the job satisfaction among blocks and within blocks are analyzed. Regarding the latter, significant differences between clusters in terms of working conditions and job satisfaction are observed (see Table 7 ). Comparing job satisfaction among blocks, significant differences are observed. Likewise, analyzing the working conditions among blocks, differences among all of them exist, with the exception of clusters 1 and 2, which show similarity in their work time quality; and between clusters 2 and 3, which show similarity in physical environment and work intensity. There is a high heterogeneity in the variables related to working conditions in the three groups identified, endorsed by medium-high effect sizes in many of the comparisons that are also higher than the effects found in the original blocks identified in the literature.

Table 7.

Inter-group differences in working conditions and job satisfaction (proposed classification). t-test. P-value (Effect size).

Variables	Group1 vs. Group2	Group1 vs. Group3	Group2 vs. Group3
Earnings	0.008 (0.208)	0.000 (0.396)	0.000 (0.769)
Prospects	0.002 (0.227)	0.000 (0.534)	0.000 (0.754)
Social environment	0.042 (0.154)	0.001 (0.193)	0.000 (0.349)
Physical environment	0.000 (0.591)	0.000 (0.523)	0.487 (0.056)
Work intensity	0.000 (0.460)	0.000 (0.577)	0.151 (0.116)
Skills and discretion	0.000 (0.299)	0.035 (0.118)	0.000 (0.420)
Work time quality	0.999 (0.000)	0.000 (0.421)	0.000 (0.440)
Job satisfaction	0.0021 (0.090)	0.0001 (0.177)	0.0511 (0.074)

¹p-value of Mann-Whitney test (four-point Likert scale).

Regarding the differences across countries within each cluster, it can be observed that there are no significant differences in job satisfaction among the countries that are part of the same block (see Table 8 ). Regarding working conditions, a high degree of homogeneity is observed. Countries in groups 2 and 3 show the greatest homogeneity in working conditions. Although there are significant differences, especially in group 1, the effect sizes are small. This is observed in the ANOVA test (except in earnings and, to a lesser extent, in prospects in group 1), and in the rest of the tests performed, as shown in Table 8.

Table 8.

Intra-group differences in working conditions and job satisfaction (proposed classification). ANOVA test. P-value (Effect size).

Variables	Group 1	Group 2	Group 3
Earnings	0.000 (0.148)	0.363 (0.015)	0.0071 (0.310)
Prospects	0.000 (0.072)	0.223 (0.020)	0.0201 (0.235)
Social environment	0.003 (0.029)	0.178 (0.023)	0.3371 (0.105)
Physical environment	0.040 (0.019)	0.189 (0.022)	0.0111 (0.222)
Intensity	0.125 (0.015)	0.148 (0.024)	0.8041 (0.025)
Skills and discretion	0.000 (0.057)	0.009 (0.051)	0.0001 (0.366)
Work time quality	0.702 (0.007)	0.070 (0.032)	0.1901 (0.132)
Job satisfaction	0.0002 (0.040)	0.1402 (0.025)	0.9923 (0.000)

¹p-value of t-test (2 countries). ² p-value of Kruskal-Wallis test (four-point Likert scale). ³ p-value of Mann-Whitney test (four-point Likert scale and 2 countries).

The workers in the three defined blocks are homogeneous in terms of characteristics such as age, gender, seniority in the company, and the percentage of self-employed people, as shown in Table 9 .

Table 9.

Working conditions and demographic characteristics in the proposed blocks.

	Groups
	1		2		3
Gender	Percentage of women
	44.2		42.6		43.6
Self-employed	Percentage of self-employment
	19.5		19.4		21.1
Variables	Mean	SD	Mean	SD	Mean	SD
Age	41.22	13.60	41.18	13.35	38.89	11.16
Seniority (years)	8.29	9.32	8.77	9.44	7.12	8.58
Monthly earnings (€)	1400.74	903.46	1586.81	853.09	1086.70	507.29
Skills and discretion	46.61	21.07	52.90	20.82	44.12	20.93
Social environment	73.12	26.41	69.00	27.70	78.16	25.53
Physical environment	85.79	11.37	78.77	13.76	79.52	13.09
Work intensity	33.77	17.75	42.16	20.12	44.48	19.97
Prospects	60.70	19.23	65.06	19.02	50.33	19.76
Work time quality	64.80	16.26	64.80	14.86	58.08	15.44

Studying the working conditions of each block, a great disparity between the salaries of groups 2 and 3 is observed. The group composed of Greece and Spain presents lower values in all variables except social environment and work intensity. Groups 1 and 2 show similar results, but working conditions are slightly more favorable in group 2. This group presents better results with respect to the rest in skills and discretion, intensity, and prospects, while group 1 shows more favorable conditions in the social environment and physical environment dimensions with respect to the rest.

5. Discussion

This research identifies a novel grouping of European countries according to the working conditions prevailing in the tourism sector. The differences among country clusters are manifested in different levels of employee satisfaction since the institutional context greatly influences working conditions, which in turn determines job satisfaction (Salvatori, 2010; Western, 1998). Despite the relevant role of the institutional context—where planning and policymaking occur—in shaping working conditions, this issue has received little attention in the literature on tourism. Studies focused on institutional context and working conditions and job satisfaction in the tourism industry are rare. According to Western (1998), working conditions are highly influenced by national regulations—and especially by labor regulations—and therefore by the institutional context. The strength of unionization becomes an important factor influencing job satisfaction because employees’ wellbeing is highly determined by salary and work intensity, among other working conditions, which are especially influenced by the levels of unionization (Pichler and Wallace, 2008). Since strong unionization in a country can lead to better working conditions, the relevance of the institutional context as an antecedent of working conditions and job satisfaction must be highlighted.

Classifying European countries according to their working conditions can set the basis for a deeper understanding of the factors that determine job satisfaction in the tourism industry in different territories. As has been concluded from the analysis, a classification of countries based on their institutional characteristics as proposed by the previous literature (e.g., Albareda et al., 2007; Brookes and Barfoot, 2005; Filella, 1991; Ronen and Shenkar, 1985; Tangian, 2008) does not group countries correctly according to working conditions and job satisfaction perceived by workers. Few differences in worker satisfaction among countries that have different institutional settings and large differences among countries of the same institutional context have been found. Similarly, countries of different institutional environments have similar working conditions, while countries of the same context present large differences in working conditions. These results point to the need to propose a new classification or clustering of European countries according to their prevailing working conditions and job satisfaction levels. Although the comparison of job satisfaction across European countries has been studied by academics, previous studies have analyzed individual countries without considering the existence of homogeneity among countries and the existence of differentiated blocks in terms of their institutional setting. This research proposes a novel classification of countries according to prevailing labor conditions in each territory—what marks differences in job satisfaction across country clusters.

One of the key aspects that determines working conditions is labor flexibility, and this depends to a large extent on institutional context (Posada-Kubissa, 2018). Tangian (2008) affirms that policies that enhance flexible employment are incompatible with achieving employment security. Carr and Chung (2014) propose that in countries where the levels of labor flexibility are high, employment security policies should be implemented to increase employees’ security. Therefore, different levels of employment protection and labor flexibility determine different social systems. Despite the EU ´s supranational government, there are differences in social systems across countries (Brewster and Hegewisch, 2017). Sapir et al. (2004) identified four social models in Europe, each emphasizing security versus flexibility to a different extent: flex-insecure, inflex-secure, inflex-insecure, and flex-secure.

According to our analysis, group 1 corresponds to two groups of inflexible countries according to Sapir’s classification: The Continental cluster (inflexible and secure: Austria, Belgium, Germany, Italy, Norway, the Netherlands, and Switzerland) and the countries included in the Anglo-Saxon block (inflexible and insecure: Ireland, Portugal, and the United Kingdom). The former are countries characterized by high income inequality, low-wage jobs, high levels of employment protection, low job security, and by early retirement pensions (Sapir et al., 2004). According to the previous characteristics and inspired by Sapir et al. (2004), we propose to call group 1 as inflexible group. According to Probst et al. (2017), this model was considered to be effective in reducing poverty but ineffective in job creation in the long term. On the contrary, the Anglo-Saxon model is characterized by low-wage jobs, low job security, and high levels of income inequality. This model was effective in creating employment opportunities but ineffective in reducing poverty.

Group 2 resulting from our analysis corresponds to the Scandinavian model (Denmark, Finland, France, and Sweden). This country cluster is characterized by a robust social security system. Although job protection is low, employment security is high in comparison to the rest of the blocks. This model enhances job creation and a high standard of living. The countries grouped in this cluster present similar levels of employment protection and low levels of inequality (Gil-Alana et al., 2019). Therefore, following Sapir et al. (2004), we propose to call this country block as flex-secure.

Group 3 resulting from our analysis, the so-called Mediterranean model (Greece and Spain), emphasizes employment protection and early retirement pensions (Probst et al., 2017). Inspired by Sapir et al. (2004), this cluster could be called flex-insecure because both countries in this group show high levels of flexibility and insecurity. According to our results, Greece and Spain show homogeneity in their working conditions. These countries experienced a deep recession after 2008, leading to an economically inferior position within Europe. They are characterized by their weak institutions and the fiscal balance programs that have been implemented by their governments following the recession. Both countries have been highly affected by prolonged austerity policies and present the highest levels of unemployment in comparison to other European countries (36.8 % in Greece and 34.9 % in Spain), according to Eurostat (2018). This can be an important factor that determines the differences found in this research in comparison with the blocks identified by the literature, which groups these two countries according to the institutional and organizational characteristics. While Filella grouped Italy, France, and Spain within the Mediterranean cluster in 1991, the socioeconomic development of each country has been different in the past decades. While France and Italy have improved their working conditions, Spain has remained among the countries with low job security and high flexibility in its labor market, which is reflected in the lowest levels of job satisfaction, showing more similarities to Greece in terms of working conditions and job satisfaction than to Italy and France.

6. Conclusion

Two main motivations led us to focus our analysis on the tourism sector: its high weight in the economy of European countries (World Tourism Organization, 2018) and its characteristics that entail high levels of precariousness (Jovanović et al., 2019). The results of the empirical analysis show that classifying countries according to their institutional setting does not properly reflect the differences in working conditions and job satisfaction across Europe. This study proposes a novel classification of European countries according to working conditions in order to understand the differences in job satisfaction in different European countries from an employee perspective. The results point to differences among countries that present similarities in their institutional context. This is observed in the higher levels of satisfaction that countries such as France, Italy, and Portugal present in comparison with Spain and Greece (all of them belonging to the same block according to previous studies). The great differences among countries that belong to the same block and the small differences in working conditions among the countries of different blocks (with the exception of the Latin cluster) lead us to posit the need to propose a novel classification of countries according to their working conditions.

Our research results show the existence of different models of working conditions in Europe that go beyond the national borders of each country. The existence of three differing working conditions models—and subsequent differences in the levels of job satisfaction—are determined not only by institutional factors, which are similar among some European countries, but by other factors that need to be further analyzed such as companies’ freedom of action in labor policies and workers’ perceptions. This follows from the results of our study, which show that the grouping of countries according to their institutional context does not correspond to the grouping of countries according to their working conditions. Therefore, it can be inferred that working conditions are not only a reflection of the institutional characteristics of the territories, but that other factors must be explored to understand the differences in working conditions and job satisfaction across Europe.

Although previous classifications of European countries according to their institutional context and the model of managing employees exist (e.g., Brewster and Tregaskis, 2003; Filella, 1991; Ignjatović and Svetlik, 2003; Nikandrou et al., 2005) that take into account different aspects such as regulatory framework, economic and legal characteristics, and the type of educational system prevailing in each country, our research highlights the need to complement these studies with the employee's perspective. Human resources policies are instruments that seek to ensure the proper functioning of organizations, but this will not be achieved if these policies do not generate job satisfaction. Hence, the relevance of complementing studies that adopt an organizational perspective with the employee's perception of their working conditions and level of job satisfaction.

The research makes several contributions to the literature. First, studies on the relationship between the institutional framework and working conditions in the tourism sector are rare. Previous research does not explore the differences between the framework in which the working conditions are developed (which is highly influenced by the institutional context where the company operates) and the labor conditions developed at the organizational level, both determining job satisfaction. Previous works that classify countries according to their institutional characteristics have only considered the framework in which working conditions are developed, ignoring that organizational management highly determines working conditions. In this vein, this study complements existing literature by proposing a novel classification of European countries based on the working conditions developed at the company level and by considering workers´ perceptions about these conditions and their job satisfaction.

On a practical level, the research shows how European countries are grouped according to workers' perceptions of their working conditions in the tourism sector. The results show that, although the institutional context is decisive in working conditions, these conditions are not determined entirely by these factors since there are territories with similar institutional settings but with substantially different working conditions. Grouping European countries according to their homogeneity in working conditions is particularly interesting for understanding international differences in job satisfaction since work satisfaction is a direct reflection of organizational policies and practices and the extent and character of institutionalized labor norms and regulations.

These results have implications for organizations and policy-makers. For organizations, assuming the freedom of movement of workers in Europe, companies can attract talented employees from different European countries if they improve their working conditions by assimilating them to the territories with higher levels of job satisfaction. For European policy-makers, interesting conclusions might be drawn from this research. To advance the EU convergence, it is necessary to homogenize the working conditions of the European countries, aiming to reach those conditions that achieve the highest degrees of job satisfaction. This will have benefits not only at the individual level, but also at the organizational and social levels. This need is especially emphasized in the uncertain context in which the tourism sector finds itself due to the COVID-19 pandemic. It is difficult to predict the structural changes that the economic crisis expected after the pandemic will generate in the tourism sector, but it is expected that demand could contract in the near future due to the economic crisis predicted by international organizations such as the International Monetary Fund (2020). The expected contraction in demand could be seen as an opportunity to create a more sustainable tourism model that prioritizes quality over quantity, a more balanced tourism model that distributes its value more equitably and fairly among the different stakeholders. Considering the fundamental role played by employees in the quality offered in the tourist service and their important contribution to business success in this sector, a model based on quality must be accompanied by better working conditions that result in greater employee wellbeing.

Despite the usefulness of this study, the results should be taken with caution due to the following methodological limitations. In the first place, job satisfaction is measured through self-perception, which can generate some bias in terms of the use of variables with an objective nature. Second, the problem of comparing countries involves the bias that is introduced regarding different variables such as salary, which cannot be compared in absolute terms without considering the cost of living, and the expectations of employees in each country. Future research could include perception variables about satisfaction with a salary instead of the salary in absolute terms to make the data comparable across countries. The classification of countries proposed by this study sets the basis for a deeper discussion on the factors—beyond the regulatory pressures that shape the institutional context—that influence working conditions. Therefore, future research could explore factors such as the culture that might be similar in each of the clusters identified and that can be determinants of job satisfaction. Finally, exploring job satisfaction in sectors different from tourism might lead to different groupings due to the specific characteristics of each sector. Therefore, future research could replicate this study in other industries.