BURNOUT AND EMPLOYEE PERFORMANCE IN HOSPITALITY INDUSTRY: THE ROLE OF SOCIAL CAPITAL

The hospitality industry is known for its labor-intensive and intense interpersonal relationships characteristic. The high rate of burnout in hospitality industry has been a crucial issue. The burnout could lead to employee performance decline. The previous studies suggested that the happy and cooperative employees tend to deliver a better performance. Since cooperative is a part of social capital, this study aims to analyse the role social capital in reducing burnout and improving employee performance. The data collection was conducted by distributing questionnaires to all of non-daily worker employees at the first-line and middle-line level in three budget hotels in the similar chained hotel group in Surabaya. We processed the data using partial least square analysis technique.The result reveals that the social capital have a significant negative influence on burnout, and a significant positive influence on employee performance. We also ascertain that burnout have a significant negative influence on employee performance.


INTRODUCTION
The hospitality industry is known as a labor-intensive industry. The intensity of the workforce in the hospitality industry has always been attributed to the irreplaceable role of personal services. Working in the hospitality industry can be tiring for employees. They face demanding work demands, complex procedures, and intensive interpersonal relationships at every stage of their working day (Birdir andTepeci, 2003 in Yirik, Oren andEkici, 2015). Moreover, employees must be responsive to serve customers who have different needs with minimum error rates. Therefore, no wonder they are easy to burnout. According to Huang and Wang (2011), burnout is a psychological syndrome consisting of: a) chronic fatigue, sleep disorders, different physical signs; b) pessimistic and negative tendencies toward colleagues and clients, feeling guilty, cornered, job dissatisfaction; c) feelings of failure and inability, loss of judgment and understanding, feeling pressed and exploited, and loss of performance.
Employees who experience burnout will certainly reduce their performance. They will tend to be passive and feel pessimistic about the completion of their work that causes them to become more depressed. They will also be vulnerable to diseases that then potentially increase their absence rates as well. Boehm and Lyubomirsky (2008) also pointed out that happy employees show higher levels of performance than unhappy employees do, because they are more sensitive to job opportunities, more involved and helping others, more confident and optimistic (Zelenski, Murphy & Jenkins, 2008).
Furthermore, El-Said (2013) argued that cooperation is one of the factors that affect and improve employee performance at the hotel. Cooperation as one dimension of social capital arises because of a certain set of values or informal norms among group members (Fukuyama, 1995, in Bhandari & Yasunobu, 2009). The research of Dai, Mao, Zhao and Matilla (2015) also reveals that social capital can improve employee performance. The social environment that has the characteristics of mutual trust, the common purpose, and cooperation will improve employee morale. The existence of such spirit will boost their performance (Shirom, 2009). It can be concluded that employees with high levels of social capital will work harder when trying to do the best for their organizations and colleagues (Hador (2016).
A number of studies have investigated the relationship between health, social capital and its components. One of which studies in Canada shows that social capital, especially trust, has a significant relationship with reducing depression levels (Sheingold, Hofmeyer & Woolcock, 2012). The phenomenon of social capital, burnout and employee performance has been extensively studied in midscale and luxury hotels (including Kuruuzum, Anafarta & Irmak, 2008;Fiksenbaum, Jeng, Koyuncu & Burke, 2010;Karatepe & Tizabi, 2011;Yirik, Oren & Ekici, 2015). However, there is still lack of research linking social capital, burnout, and employee performance in budget-class hotels. For that reason, this study will investigate the impact of social capital on burn out and employee performance in three budget hotels in the similar chained hotel group in Surabaya. The research will be conducted on a number of non-daily worker employees at the first-line and middle-line level. The non-daily worker is an employee who has passed a probationary period or has met the criteria as a contract employee, so his/her relationship with a co-worker is considered strong enough. The employees must be on the first line and middle line level, because only at that level employees have co-workers with the same level of office. Nahapiet and Ghoshal (1998), the experts in social capital, define social capital as a number of actual and potential resources embedded in, through, and derived from a network of relationships owned by an individual or a social unit. Their original writing becomes the foundation of many further researchs. Unlike human capital, which is a combination of attributes, skills, and experience of a person, social capital is values and benefits, actual and potential, generated from social interactions of the person (Santarelli & Tran, 2012). Social capital cannot be separated in relationships between individuals. As a set of resources derived from relationships, social capital has many different attributes. Nahapiet and Ghoshal (1998) divide social capital into three dimensions, namely: structural social capital, relational social capital, and cognitive social capital. We use the social capital dimensions according to Nahapiet and Ghoshal (1998), because this study analyzes the influence of social capital at the individual level within the organization.

Social Capital
The structural social capital is an overall form of relationship between social actors Nahapiet and Ghoshal (1998). This term describes the configuration of impersonal relationships of people or units. According to Nahapiet and Ghoshal (1998), this dimension refers to the organizational structure, the pattern of connections between individuals, and the relationships that make up the organization's network. This dimension has the meaning that a person's position in the interaction structure will give him certain advantages (Nahapiet & Ghoshal, 1998), such as the ease of finding jobs, obtaining information, or accessing resources (Tsai & Ghoshal, 1998).
Relational social capital refers to assets rooted in the relationship, such as trust and reliability to be trusted (Tsai & Ghoshal, 1998). Trust is the attribute of a relationship, while reliability for trustworthiness is an attribute of the individuals involved in the relationship (Barney & Hansen, 1994in Tsai & Ghoshal, 1998. This dimension supports performance cohesion because it reflects mutual trust, togetherness, and caring. This dimension is an asset that is created and grows in relationships among members of the organization that include beliefs, norms and sanctions, obligations, expectations, and identification (Nahapiet & Ghoshal, 1998).
The cognitive social capital dimension according to Nahapiet and Ghoshal (1998) refers to resources that provide representations, interpretations, and systems of shared meanings. This dimension is manifested in attributes such as shared codes or shared paradigms that facilitate a common understanding of common goals and appropriate ways of acting within social systems. This general understanding can be done by collectivity as a resource. This is added in attributes such as shared vision or equality of values that facilitate individual and collective action and shared understanding of appropriate action and collective goals. The cognitive dimension includes attributes such as shared norms, action codes, and convergence of views (Zhang et al, 2011). Shared values and visions can foster the development of relationships for mutual trust. Members of the organization with collective goals and values will tend to trust each other, because they can expect them to work together to achieve collective goals and will not be impeded or imposed by other members for pursuing self-interest (Tsai & Ghoshal, 1998).

Employee Performance
According to Hafeez and Akbar (2015), performance can be interpreted as the achievement of certain tasks that are measured based on predetermined standards or identified with the accuracy, completeness, financing and speed. According to Bakker and Schaufeli (2008), every employee in the organization is required to provide a positive contribution through good performance, given the performance of the organization depend on the performance of its employees. Employee performance is very important because it will reflect the organization's performance. Salleh, Mishaliny and Haryanni (2012) reveal that employees play important role in ensuring effective and efficient implementation of organizational policies and programs. Mathis and Jackson (2010) suggest four indicators to measure the employee performance. We use these indicators because one of the indicators is the cooperation with colleagues, so it has a correlation with social capital. In addition, within the working environment of the hospitality industry, the ability to work together is one of the important measurement to assess the employee performance. These indicators are as follows: 1) Quality of work; measured from the employee's perception of the quality of work produced and the perfection of tasks compared to the skills and abilities of employees. 2) Quantity of work; measured from the employee's perception of the number of activities assigned and the results. The quantity measurement involves calculating the output of the process or the execution of the activity. 3) Work time; measured from the employee's perception of time for an activity accomplishment compared to the appointed time and the ability to maximize the time available. 4) Cooperation with colleagues; measured from the ability of employees to cooperate with colleagues and the environment. The ability to work together can create cohesiveness so that it can improve the teamwork sense among employees.

Burnout
Burnout is a prolonged stress, a demand in the workplace that burdens or exceeds the resources owned by individuals (Buick & Thomas, 2001, in Lu & Gursoy, 2013. According to Brill (1984), stress refers to a transient adaptation process and is accompanied by mental and physical symptoms, whereas burnout refers to a disturbance in adaptation accompanied by chronic functional impairment (in Buick & Thomas, 2001, in Lu & Gursoy, 2013. The stress that causes this burnout includes high tension in work, low social support, exposure to workplace violence and intimidation, night shift work demands, high demands at work, poor work organization, ambiguity in decision making in critical situations based on inadequate information (Farzianpour, Fouroshani, Mohamadi & Hosseini, 2013). Burnout is not a personal matter. It is a social or environmental issue related to one's work (Beckstead, 2002). Baron and Greenberg (2003) suggest four indicators of burnout, while Maslach, Jackson and Leiter (1997) propose three indicators only. The difference is that Maslach, Jackson and Leiter regard physical exhaustion as one of the impacts of burnout, while Baron and Greenberg consider physical exhaustion to be a form of burnout. In this study, assume that physical exhaustion is one form of burnout, because employees who are constantly exposed to stress will be susceptible to disease and poor lifestyle tendencies such as decreased appetite and insomnia. Physical exhaustion does not always appear after a person experiences all types of burnout. It can be felt when employees feel exploited and perceive their responsibility exceeds the resources they possess.

The Relationship between Social Capital and Employee Performance
An atmosphere built on trust, shared values and beliefs can help people to collaborate and make them easier to assess their working conditions by reducing insecurity, uncertainty, and disorientation. These conditions can also improve their performance (Ommen et al., 2009). Social capital can affect the quality of service and output. The existence of social capital between employees will increase their morale to work better. Good social relations among employees will create a comfortable working environment. This work environment is created because of the common vision and goals among employees that support the performance of employees to cooperate and achieve common goals. Cooperation can be created if the trust is inherent in the parties who trust and believe in the given task. Support from colleagues or superiors also encourage employees to work well. This support is the result of trust and a close network between the parties concerned. Therefore, we propose the following hypothesis: Hypothesis 1: Social capital positively influences employee performance. Farzianpour et al. (2013) found that social capital has a significant and inverse relationship to burnout which signifies the importance of social capital role of employees in an organization. Support from colleagues can help employees to cope with stress and reduce the chances of experiencing burnout. Corporate custom as a form of norm in social capital is also able to reduce burnout caused by conflict. The manners ignorance can create a less conducive working atmosphere. Moreover, the neglect of etiquette can trigger personal conflicts that are difficult to overcome (Boyas & Wind, 2010). Such conflicts can cause emotional exhaustion and depersonalization, which is a reflection of burnout. Hypothesis 2: Social capital negatively influences burnout.

The Relationship between Burnout and Employee Performance
Burnout can negatively affect employee attitudes and lead to negative behaviors, such as low work involvement, performance decline, and increased turnover intentions. The negative effects of burnout on employee performance may be lowered effectiveness, work absenteeism, decreased service quality, loss of interest in the organization, family and marital problems, alcohol and drugs consumption, depression and even suicidal tendencies. Therefore, recognizing this syndrome and its effects and preventing the occurrence of this syndrome is very important (Farzianpour et al., 2013, andYirik, Oren &Ekici, 2015). Hypothesis 3: Burnout negatively influences employee performance.

RESEARCH METHOD
This study aims to determine the effect of social capital on burnout and employee performance. The population of this study is 51 non-daily worker who are in the first and middle level of three budget hotels in the similar chained hotel group in Surabaya. The first-line employees are employees who hold the position of ordinary staff, while middle-line employees are employees who have the position of supervisor and manager or head of department. Due to the small population, we use saturated sampling techniques.
The exogenous variable in this research is social capital (SC). In this study, social capital is defined as a collective asset in the form of norms, beliefs, networks that are shared together that lead to cooperation and collective action for mutual benefit. The social capital dimension used is adapted from Nahapiet and Ghoshal (1998) to measure how strong social capital among hotel employees is. These dimensions are structural social capital, relational social capital, and cognitive social capital. One example of structural social capital indicator is "I feel that the work team facilitated me to participate in work activities." The endogenous variables in this study are burnout and employee performance. We adapt burnout indicators from Baron and Greenberg (2003) which consist of four dimensions. One example of burnout indicator is "In the past month or so, I find it is hard to concentrate". The employee performance indicators adapted from Mathis and Jackson (2002) consisting of four indicators. These indicators are quantified as follows: "I am able to complete the job on time" In this study, we collected data by distributing questionnaires using Likert scale with scale of 1 (strongly disagree) to 4 (strongly agree). We used a four categories likert scale to avoid respondents' tendency to give neutral answer. Afterward, the data is analysed using Partial Least Square (PLS) method. We apply Partial Least Square (PLS) because it requires relatively small data and more flexible assumption requirement.
To test the validity and reliability of reflective contructs, we utilize benchmarks, as stated in Table 1. Meanwhile, to examine the validity of the formative construct, the evaluation of measurement model is based on the significance of T-Statistics of formative construct. Hence, the validity and reliability test is not needed. To get the T-Statistics value through the bootstrapping process, the T-Statistics value should be> 1.96 to be valid. The structural model is evaluated using R-square (R²) for dependent constructs and T-Statistics for significance of influence. The criteria is R 2 > 0.3 (very weak), 0.3 < R 2 <0.5 (weak), and 0.5 < R 2 <0.7 (moderate), and R 2 > 0.7 (strong) (Moore, Notz & Flinger, 2013).
In addition to the R-square values, the PLS model is also evaluated using the T-Statistics score to measure the significance of the latent construct influence on other constructs. The size of the significance of T-Statistics should be more than 1.96. The level of confidence used is 95%, so the level of precision or inaccuracy limit α = 5% and yielded a T-Statistics score of 1.96.

FINDINGS
Based on the questionnaire that have been filled by 51 respondents in three budget hotels in Surabaya, we notice that the majority of respondents are in the age range 26 -30 years old (41%), men (53%), married (71%), in the middle-line level management (55%), length of work > 1 year (94%) and at Front Office department (32%). Figure 1 shows the result of the outer model that has been constructed and processed. The convergent validity can only measure variables with reflective indicators only, i.e. burnout and employee performance. All loading factors that relates among indicators, and between indicators with variables, have value > 0.4. Thus, we can conclude that the indicators and the variables are valid.

Figure 1 Outer Model
Unlike burnout and employee performance, the social capital variable has formative dimensions. Hence, we do not assess its validity through loading factor in outer model, but through T-statistics at inner stage model. On the other hand, the relationship between the indicator statement and its dimensions is reflective, so that it can be measured through the loading factor value. In this study, the structural social capital dimension consists of six indicators, relational social capital dimension consists of seven indicators, and cognitive social capital dimension consists of six indicators. All of them have loading factor values > 0.4. Thus, all indicators are valid. The test results show that all dimensions of social capital, employee performance, and burnout have AVE value > 0.5. This means that all reflective variables with indicators along with the three dimensions of social capital meet the standard value of convergent validity. The value of AVE social capital is only 0.297, but this should not be considered. Since the social capital is a formative variable, the AVE value is not required.
Based on the results of discriminant validity test, all values of cross loading construct associated with the indicator is higher than the value of other constructs. Thus, all constructs in this study have met the discriminant validity standart. The result of data processing shows that all constructs have composite reliability value > 0.7 and Cronbach's alpha > 0.7. Hence, we can confirm that all constructs are reliable.
To asses the reliability of the formative variable, we tested the inner model with bootstrapping and obtained the formative T-statistics value of the construct. In this study, the formative construct is social capital with its three dimensions. All of social capital dimensions, namely: structural, relational, and cognitive social capital have T-statistics > 1.96. Thus, we can conclude that these three dimensions of social capital are valid. The next step after testing the outer model is to test the inner model. We evaluate the value of R 2 to find out the predictive power of the effects obtained by the endogenous constructs of the exogenous constructs that influence it. Figure 2 shows the inner model that has been constructed and processed to obtain the Tstatistics value of each construct.  Table 2 shows that social capital variable has R 2 = 0.999, meaning that the three dimensions of social capital have prediction influence of 99.9% to the variable that can be categorized strong. All burnout dimensions have R 2 > 0.7, which means that each indicators has a strong influence on its dimension. The employee performance variable obtained R 2 = 0.276 which is categorized very weak. It means that the variable social capital and burnout influence employee performance just as much 27.6%. The work quality and working time indicators are categorized strong because the value of R 2 > 0.7. As with the indicator of quantity of work and cooperation are categorized moderate. The burnout variable obtained R 2 equal to 0,129, meaning social capital variable predicted to affect burnout as much as 12.9% and categorized very weak.

Figure 2 Result of Inner Model
We execute the hypothesis test to find out the influence and significance between variables. Table 3 shows that all of T-statistics are above 1.96. Subsequently, we can state that all variable relationships are significant.

The Influence of Social Capital on Employee Performance
In this study, the social capital influences employee performance positively and significantly (T-statistics = 2.156). The influence is positive, meaning that the increase of social capital will lead to employee performance increase. This result is in line with Hador (2016) which reveals that strong social capital in the workplace will make employees feel better, more energized, and eventually there will be an improvement in employee performance.
The results of the questionnaire show that respondents tend to answer, "Strongly agree" on the social structural dimension indicators. This means that they have a very good relationship. In the hotel work environment, there is a clear organizational structure and an interconnected work team. The work environment in the hotel requires high intensity face-to-face interaction with colleagues, even with colleagues from different departments.
As Tsai and Ghoshal (1998) proposed, the structural social capital can stimulate trust that represents relational capital. The people who frequently interacts will create a more concrete relationship of trust. In addition, they also argued that social interaction facilitates the shared goals and values forming within the organization. The daily routine activities, such as morning briefings that bring together the managers of each department to evaluate the work teams, will generate the familiarity and acquaintance between employees. It affects the personal relationships between employees, which based on mutual trust and norms established in their relationships. Therefore, it is not surprising that the dimensions of relational social capital and cognitive social capital obtained a high mean value.

The Influence of Social Capital on Burnout
The social capital has negative and significant influence on burnout (original sample= -0.293, T-statistics = 2.041). Since the majority of the respondents choose "disagree" on the burnout indicators, it can be inferred that the issue of burnout in the work environment is minor. One of possible explanations of this minor burnout is a high level of social capital. This is align with Farzianpour et al., (2013) in their research in teaching hospitals in Iran. The result shows that social capital, consists of the mutual trust, the ability to create informal social relationships, generosity and volunteerism; variations in interaction, friendship, and leadership; and community involvement, have a significant inverse relationship to burnout.
However, the burnout variable has a R 2 value of 0.129, which is categorized as very weak. This means that although social capital has a significant influence on burnout, the predicted influence obtained by burnout from social capital is very weak, i.e. 12.9% only. There are 87.1% influence of other variables affected burnout that are not examined in this study.
The minor burnout probably because employees already have satisfactory work experience. This is supported by Farzianpour et al., (2013) studies in Iran teaching hospitals. They reveal that the higher the nurse's experience, the less likely the nurse had burnout. In addition, Ang et al. (2016) found that the older nurses and the one with longer working periods had less potential for burnout than those with younger age.
The demographic factors such as age, gender and marital status also potential to influence the burnout. The more mature employees have a lower possibility for burnout. Whereas, women and unmarried employees have a higher tendency to burnout than men do. This is in line with Buick andThomas (2001, in Lu &Gursoy, 2013) who found that the younger, female, and unmarried employees are more vulnerable to burnout. Ang et al. (2016) also found that demographic factors, such as the age and races, influence burnout. Regarding the demographic profile of respondents dominated by a relatively mature, male, and married employee, we suggest that the demographic factor is a potential contributor to the minor burnout.

The Influence of Burnout on Employee Performance
The third hypothesis stating that burnout has an influence on employee performance is accepted (T-statistics value = 2.155). The value indicates that burnout has a significant effect on employee performance. The effect of burnout on employee performance is negative, as shown in coefficient value is equal to -0.342. This is in line with Kuruuzum, Anafarta and Irmak (2008), that burnout can lead to decreased work performance, the emergence of a desire to quit the job, work absenteeism, family problems, decreased self-esteem, difficulty in concentrating, social disengagement, adverse physical symptoms (sleep disturbances, headaches, etc.), alcohol and drugs consumption, psychological disorders (anger, depression and apathy). Similarly, Farzianpour et al., (2013) point out that the negative effects of burnout on employee performance could be in the form of reduced effectiveness, absenteeism, decreased patient satisfaction, family and marital problems, alcohol consumption and drugs, depression and even suicidal tendencies.
In the hospitality industry, burnout may possibly trigger poor customer services and potential error increase. This is supported by Ari and Bal (2008), that the consequences of burnout include negative behavior toward customers, reduced service quality, the potential for more frequent errors, loss of interest in the organization and work, loss of creativity, job dissatisfaction, poor performance and professional decline in work, procrastinating assignments, and absence (in Yirik, Oren and Ekici, 2015).