EXECUTIVES’ FRINGE BENEFITS IN LANGKAWI HOTELS

An important element of human resources, compensation is an effective tool to attract and retain skilled employees. Despite its crucial importance the literature concerning executive compensation in the hotel industry is very few especially those focusing on fringe benefits. To fill the gap in the hotel human resources literature, this study assessed the level of fringe benefits provided to executives in the hotel industry. The purpose of this empirical study was to gain information regarding executives’ compensation in the industry. Stratified random sampling technique was used to ensure that the findings represented hotels of various star-ratings. A total of 250 questionnaires were distributed to the ratio of 6:3:4 to three-, four- and five-star hotels and 104 were returned. The findings revealed that there is a statistically significant difference in the fringe benefits given between three -, four - and five-star hotels, and the star rating of hotels positively correlates with the level of fringe benefits provided to hotel executives.


Introduction
Compensation is an important component of human resource management. It can serve as a tool to attract, motivate and retain skilled workers (Hayes & Ninemier, 2008). There is a consensus that compensation is one of the strongest determinants of attitude and behaviour (Ibrahim & Boerhaneoddin, 2010;Kusluvan, Kusluvan, Ilhan, & Buyruk, 2010;Resurreccion, 2012). Compensation influences positive attitude and behaviour such as motivation, job satisfaction and organisational commitment and these behaviours reduce employee turnover and enhances business profitability (Danish & Usman, 2010;Kusluvan et al., 2010;Nankervis, 2000;Resurreccion, 2012;Sun, Aryee, & Law, 2007). Compensation can be defined as the combination of pay and rewards to employees (Dessler, Giffiths, Walker, & Williams, 1999). Compensation consists of direct financial payments such as salaries and indirect payment or fringe benefits (Hayes & Ninemeier, 2008;Tesone, 2005).
Literature concerning executive compensation in the hospitality industry is very few (Guillet, Kucukusta, & Xiao, 2012). Moreover, very few of these studies include information on fringe benefits (Ahmad, Scott, & Solnet, 2010b;Johnson, 1983). Studies on fringe benefits have been neglected even in the general management field of research (Johnson, 1983;Williams, 1992). Recent literature on fringe benefits is scarce. Johnson (1983) provided detailed information regarding fringe benefits in the hotel industry but the study was conducted almost three decades ago. As for Namasivayam, Maio and Zhao (2007), the fringe benefits examined in the study were categorised into five groups and they were not specifically listed as in Johnson (1983).
The lack of literature on fringe benefits is more critical in the Malaysian hotel industry. To fill the gap in the hotel industry literature, Ahmad et al. (2010b) assessed the fringe benefits provided to executives and non-executives in Langkawi fivestar hotels and discovered that the fringe benefits provided do not only vary between jobs but also vary between hotels. As with other studies, the study has several limitations. It only assessed the compensation practice of five-star hotels, and the study was done qualitatively. Thus, the findings cannot be generalised and the information on fringe benefits in hotels of other categories remains bleak.
Due to the gaps in the literature, a quantitative study to assess the fringe benefits provided to hotel managers would be beneficial. Furthermore, employees at executive level are more likely to remain with their employers for compensation packages that include fringe benefits rather than primarily monetary compensation (Cheng & Brown, 1998). Including hotels from various star-ratings in the research enables the gathering of more information regarding the variation in the hotels' compensation practices. Due to time and financial constraints, the study in Malaysia is focused on Langkawi island. Langkawi is a duty-free island and it is very near Thailand. The economy of the island is based on tourism activities, paddy plantation, rubber cultivation and fisheries. Langkawi was given World Geopark status on June 2007 by UNESCO (The Star Online, 2007). The annual tourist arrival has increased from 1.84 million in 2005 to 2.82 million in 2012 (LADA website). The tourists' destination is selected because it has a high number of three-, four-and five-star hotels.
The objective of this study is to assess the extent of fringe benefits given to hotel executives in three-, four-and five-star hotels and to examine the relationship between the fringe benefits allocation and the hotel's star rating. This study attempts to answer two research questions: "What is the extent of fringe benefits allocation to hotel executives?" and "Does the allocation of fringe benefits correlate with the hotel's star rating?" In this study, an executive is defined as the individual who is accountable for the work of at least one employee (Nash, 1980) and is categorised as a management employee instead of the "rank and file" or non-managerial employees by the respective hotels (Ahmad et al., 2010a). Excluded in the definition are the supervisory-level employees.

Definition of Fringe Benefit
Fringe benefit refers to rewards and perquisites given for organizational membership (Hayes & Ninemeier, 2008;Tesone, 2005). Perquisite, commonly referred to as perks is the extra reward given for executives' organizational membership (Tesone, 2005). Examples of perquisites given in Malaysian five-star hotels are free meals in the hotel's restaurant and free accommodation in the executive staff quarters (Ahmad et al., 2010b). Johnson (1983) listed life assurance, medical insurance, loan schemes, pension schemes, medical facilities, relocation expenses, personal accident insurance, company car, bonus schemes, clothing provision, sickness benefit, sports and social facilities, profit-sharing schemes, holiday entitlement, food at work, free transport to and from work, and company housing as fringe benefits. In addition to these fringe benefits, other benefits listed in Ahmad et al. (2010b) are paid sick leave, paid annual leave, free laundry, manager of the month and year award, birthday celebration and staff parties (Ahmad et al., 2010b). All these benefits, including perquisites are addressed as fringe benefits in this study. In other words, this study defines fringe benefits as any form of compensation provided by the organisation other than salary.

The Significance of Fringe Benefits
The rise in medical costs and other costs of living increases employees' awareness and recognition of fringe benefits value and this is supported by the discovery of a positive relationship between the percentages of total compensation devoted to benefits and the number of job applicants (Williams, 1992). In addition to attracting applicants for employment, past studies discovered that fringe benefits were important for employee motivation (Danish & Usman, 2010), retention (Carraher, 2011;Nankervis, 2000), organisational performance (Namasivayam et al., 2007) and organisational competitiveness (Resurreccion, 2012).
Fringe benefits make the working conditions more attractive and it helps retain experienced and competent employees (Carraher, 2011;Nankervis, 2000). Examples of employee retention through fringe benefits were evident in Carraher (2011) and Ahmad, Solnet and Scott (2010a). While the former was a quantitative study in the service industries from three countries in the Baltic region, Ahmad et al. (2010a) was a qualitative study with a small sample from the hotel industry in Malaysia. Carraher (2011) discovered that attitudes towards fringe benefits were a meaningful predictor of employees' and entrepreneurs' turnover and it had the greatest incremental contribution compared to salary and pay satisfaction. The author concluded that both, salary and fringe benefits were important. Salary was important for attracting employees and fringe benefits were important for retaining them. In Ahmad et al. (2010a), a non-managerial employee was quoted saying "This is the first and the last hotel that I'll work with. This is because I am old. If I go to another hotel I can't get the retirement benefit (pension scheme) that I'll get here" (p. 212). Only employees who remained with the same organisation for more than 10 years were entitled for the pension scheme or retirement benefit mentioned in the study. In other words, it was a long-term service award. However, it is not clear if executives were also entitled for the award. In addition to employee retention, fringe benefits enhanced employee productivity and job satisfaction and they were important especially for occupations with fewer chances for promotion (Cheng & Brown, 1998;Nankervis, 2000).
A study of various industries in Pakistan discovered that satisfaction with compensation, including the amount of fringe benefits received, were positively correlated with work motivation (Danish & Usman, 2010). Based on their findings, Danish and Usman (2010) suggested that managers and policy makers increased direct financial payments and fringe benefits on periodic basis and special occasions to keep employees motivated.
Fringe benefit has an effect on organisational performance (Namasivayam et al., 2007). Based on a sample of 1223 hotels in the US, Namasivayam et al. (2007) discovered that in the case of management employees, salary fully mediated the relationship between fringe benefits and organisational performance. The study also discovered that executives were more likely to be motivated by salary while non-executives were more likely to be motivated by a combination of salary and fringe benefits. The fringe benefits examined in their study were medical benefits, employee support, employee assistance programmes, retirement pay and insurance. Despite their findings, it was not evident that executives did not appreciate the value of fringe benefits especially when the executives were working in developing countries that paid a lower salary (Nankervis, 2000).
A study of compensation in various industries in the Philippines found that employee benefits were significant predictors of organisational competitiveness (Resurreccion, 2012). In the study, employee benefits comprised of flexible benefits that were customised to the diverse needs of the employees and could be converted to cash, retirement package, on-site personal services such as banking services and dry cleaners, and wellness programmes. The author concluded that employees were more motivated to perform when their fringe benefits allowed flexibility and convenience. In addition, the author suggested providing attractive retirement packages and profit-sharing schemes to improve the overall ability of fringe benefits to drive organisational competitiveness.
According to Johnson (1983), fringe benefits such as paid sick leave and holidays were also given in other industries. These fringe benefits were regarded as employees' rights and therefore did not offer comparative advantage. Johnson (1983) further stated that fringe benefits that were highly valued by employees were free accommodation and long-service awards because they were not given to many employees.
The increasing importance of fringe benefits is in line with the rise of the cost of living. Based on previous empirical studies, it is evident that a study focusing on fringe benefits is important because it enhances employee motivation, retention, organisational performance and organisational competitiveness.

The Allocation of Fringe Benefits
The allocation of fringe benefits differed based on employees' position and the size of the hotels (Johnson, 1983;Kline, & Yu-Chin, 2007). Johnson (1983) discovered that unit and departmental managers were provided with the greatest range of benefits and this was followed by the chefs and receptionists, waiters, chambermaids and bar staff, and finally, the porters. The compensation gap between the management and the nonmanagement employees such as the porters, chambermaids and waiters was consistent with a number of more recent literature (Hooi, 2006;Kline & Yu-Chin, 2007).
Organisation size and segment affect compensation practices. Many studies have confirmed that organisation size positively influenced managers compensation (Guillet, et al., 2012;Johnson, 1983;Kline & Yu-Chin, 2007). Organisation size is measured by sales, assets and number of rooms and employees (Nash, 1980;Kline &Yu-Chin, 2007). Besides categorizing hotels into sizes based on their room numbers, Kline and Yu-Chin (2007) divided hotel organisations into six segments based on the level of services provided. These are limited service-budget, limited service-commercial, limited service-deluxe, full service-moderate, full service-commercial and full service-luxury. According to Kline and Yu-Chin (2007), large, full-service and luxury hotels usually compensated employees better compared to the full service-moderate and the limited service hotels. This is consistent with Johnson (1983) who stated that small private hotels provided a greater range of benefits compared to large hotel companies.
Large and full-service hotels were forced to offer attractive compensation packages to attract and retain employees with the skill and competency required in serving its target market (Kline & Yu-Chin, 2007;Nankervis, 2000). The work in large, full-service and luxury hotels was more complex and therefore, required more experienced and competent individuals (Guillet et al., 2012;Kline & Yu-Chin, 2007). These organizations offered a more attractive compensation package compared to their competitors to attract people especially those from other hotels (Collins & Smith, 2006). This is an effective strategy to acquire experienced, skilled and competent staff especially in a tight labour market and it is known as job-hopping for employees (Cheng & Brown, 1998). The fringe benefits given by five-star hotels in Malaysia to executives are insurance and medical coverage, paid sick leave, paid annual leave, free meals in the staff cafeteria or the hotel's restaurant, staff discounts, free uniforms, transportation to and from work, free accommodation in the staff quarters, free laundry, manager of the month and year award, staff parties and birthday parties (Ahmad et al., 2010b).
In Malaysia, besides the information provided in Ahmad et al. (2010b) little is known about the fringe benefit allocation in Malaysian hotels. Fringe benefits were neglected in Hooi's (2006) study. Based on Ahmad et al. (2010b), only nonexecutives were given pension schemes or the long-service awards. Most of the non-executives in five-star hotels are union members and rewarding employees with pension schemes was one of the Collective Agreement requirements. Due to the qualitative nature of the study by Ahmad et al. (2010b), it is not clear if none of the executives in Malaysian hotels are rewarded with pension schemes or whether all executives in Malaysian hotels, regardless of the hotel categories, are given free accommodation. Furthermore, different from four-and five-star hotels, study on three-star hotels in Malaysia is scarce. Based on existing literature on compensation, the following were hypothesised: Ha1: The extent of fringe benefits allocated is different for various categories of hotels. Ha 2: Hotel's star rating positively correlates with the amount of fringe benefit allocation.

Research Methodology
This research assessed the fringe benefits given to managers in Langkawi hotels and compared the benefits given in the three-, four-and fivestar hotels. The elements of fringe benefits were insurance coverage, medical coverage, loan schemes, pension schemes, relocation allowance, personal accident insurance, company car, bonus schemes, paid sick leave, paid annual leave, holiday entitlement, clothing provision, free transport to and from work, free meals, free laundry services, staff discounts, sports and social facilities, profit-sharing schemes, free accommodation, manager of the month and year award, birthday celebration and staff party.
Self-administered questionnaires were used to collect data from the research sample. Before the actual data collection, a pilot study was carried out to test the research instrument's reliability and validity and to improve the quality of the questionnaire by identifying and excluding potential problems. The pilot study was conducted by distributing questionnaires to 36 hotel managers from three-, four-and five-star hotels. Cronbach's coefficient alpha was used to examine the internal consistency and the reliability of the instrument. To ensure content validity, three experts in the field were consulted. These experts were hotel managers. Improvement was made to the instrument as advised by the experts.

Population and Sample
The unit of analysis for this research was individuals. The population was the total number of hotel executives in Langkawi island. This research used the stratified random sampling technique to ensure that the findings represented three-, four-and five-star hotels. The population and the sample were identified based on organization instead of individuals because the hotel executives' name list could not be identified easily. Stratified random sampling, a type of probability sampling was used for this study to allow for generalisation of the findings.
The stratified random sampling of this study had two stages and it was done based on Sekaran (1992). The first stage was selecting the stratum to be surveyed based on the hotels' star rating and these were three-, four-and five-star hotels. Managers working in hotels with lower star rating were not selected for this study because these lodging establishments were usually owner operated. Hotels with less than 100 rooms were also excluded from this study because they usually employed very few executives.
At the next stage, hotels from each stratum were randomly selected from the accommodationlist published in the Langkawi Development Authority website. Based on the list, the identified ratio of three-, four-and five-star hotels in Langkawi was 12:6:8. The ratio of 6:3:4 hotels were randomly selected by drawing from a bowl.
The sample size of 50% of the population was sufficient to represent the population. The sample and the population were identified based on the number of hotels instead of hotel executives due to the difficulty of obtaining the statistics of the hotel executives working in Langkawi. Such statistics was not readily available due to the rapid change of hotel executives' employment from one organisation to another. The ratio of questionnaires distributed in each type of hotel was 15:20:25 and the ratio of the total number of questionnaires distributed was 90:60:100. The number of questionnaires distributed was based on the approximate number of executives in each type of hotel. Three-star hotels usually had fewer executives compared to four and five-star hotels. The questionnaires were distributed with the help of the hotels' human resource manager.

Instrument Design
A self-administered questionnaire was developed based on previous studies and theories related to fringe benefits. All of the questions were close-ended. The questionnaire was divided into two sections, A and B. The questionnaire was developed in the English language due to the importance and the wide usage of the language in the Malaysian hotel industry.
Section A consisted of socio-demographic questions to gain general information about the respondents' characteristics such as their gender, age, educational level, marital status, occupation, education and income level and the star rating of their organisation. Section B measured the extent of fringe benefits granted to executives. It had 22 items that were rated using a 3-point Likert scale; "does not apply", "inconsistent" and "apply completely". The 3-point Likert scale was used because based on Ahmad et al. (2010b), most respondents either answered "we don't have that", "sometimes we are given the benefit" or "we have that" when asked if they were given certain fringe benefits. This implied that the fringe benefits were either: not given at all, inconsistently given or consistently given.
While "does not apply" referred to "not given at all", "inconsistent" referred to "inconsistently given" and "applies completely" referred to "consistently given".
Fringe benefit items in the questionnaire were adapted from Johnson (1983). Specifically, sixteen items were taken from Johnson (1983) and included were six items from the findings of Ahmad et al. (2010b) that was absent from Johnson's study. The sixteen items taken from Johnson (1983) were insurance coverage, medical coverage, paid sick leave, loan schemes, pension schemes, relocation allowance, personal accident insurance, company car, bonus schemes, holiday entitlement, clothing provision, free transport to and from work, free meals, sports and social facilities, profit-sharing schemes and free accommodation. The six items taken from Ahmad et al. (2010b) were free laundry services, staff discount, manager of the month and year award, birthday celebration, staff party and paid annual leave.

Data Collection
Out of the 250 questionnaires distributed, 104 were returned, showing a response rate of 42% which was low but sufficient. The ratio of the questionnaires returned by the three-, four-and five-star hotels was 32:51:21. Several efforts were taken to improve the response rate. Phone calls were made to the Human Resource managers to remind them to distribute the questionnaires and to inquire whether the questionnaires had been returned. Effort to collect the data was taken with care to avoid nnoying the Human Resource managers. Most of the Human Resource managers gave full cooperation and all except one were helpful when they were asked for a closeended question interview. Despite the support received from the Human Resource managers, the response rate was low except in the fourstar hotels. Difficulty in getting the respondents to return the questionnaires lengthened the data collection period to more than three months.

Data Analysis
The data collected in this study was analysed using the Statistical Package for Social Science version 19.0 for Windows software programme. The analyses of data involved three stages. Firstly, Cronbach's coefficient alpha was used to examine the internal consistency and the reliability of the instrument. Reliabilities of less than 0.6 are considered to be poor, those in the 0.7 range are acceptable and those over 0.8 are good (Sekaran, 1992). Secondly, the Kruskal-Wallis Test was used to determine the differences between the three groups. Finally, Spearman correlation was used to examine the correlation between the hotel's star rating and the fringe benefits. A non-parametric test was used instead of parametric because the data was not normally distributed.

Respondents' Profile
Most of the respondents were female (58.7%). More than half (55.8%) of the respondents were between 31 to 40 years old. Most (29.8%) of the respondents were high school graduates and only 3.8% had Master's degrees. Most of the respondents were married (74%). The majority of the respondents were middle managers (39.4%) and this was followed by junior managers (34.6%), senior managers (25%) and a director (1%). Most respondents earned between RM2,001 to RM4,000 per month (51%). Only 1.9% were paid between RM8,001 to RM10,000 and RM10,001 and over 32.7% were paid RM2,000 and less, 9.6% between RM4,001 to RM6,000, and 2.9% between RM6,001 to RM8,000. The ratio of respondents' percentage from the three-, four-and five-star hotels was 31:49:20. The profiles of respondents are displayed in Table 1.
The fringe benefit with the highest percentage was holiday entitlement. It was given to all executives in three-and five-star hotels. In four-star hotels, the percentage that answered "apply completely" was very high (99%). Another 1% answered "inconsistent". The fringe benefit with the lowest percentage was profit-sharing schemes. For this fringe benefit, 1% answered "apply completely", 1% answered "inconsistent" and 98.1% answered "does not apply". The descriptive analysis of fringe benefits given in Langkawi's three-, four -and five-star hotels is displayed in month and year award (x 2 =36.108, P=0.000), birthday celebration (x 2 =34.123, P=000), personal accident insurance (x 2 =32.166, P=0.000) and sports and social facilities (x 2 =30.638, P=0.000).
Fringe benefits with insignificant differences were medical coverage, paid sick leave, paid annual leave, company car, bonus schemes, holiday entitlement, clothing provision, free meals and free accommodation. The comparison of fringe benefits given in three-, four-and fivestar hotels is displayed in Table 3.
celebration (0.567), personal accident insurance (r=0.551), manager of the month and year award (r=0.550), sports and social facilities (r=0.543), loan schemes (r=0.487) and relocation allowance (r=0.445). Other fringe benefits had a weak correlation with the hotels' star rating. The fringe benefit that did not have significant correlation with the hotels' star rating was free transport to and from work. The correlation between three-, four-and five-star hotels with fringe benefits is displayed in Table 4. Spearman's correlation test partially confirms that hotels' star rating positively correlates with the level of fringe benefits provided to executives (second hypothesis). According to Dancey and Reidy (2004), the score of 0.1 to 0.3 show weak correlation, 0.4 to 0.6 are moderate and 0.7 and above are high.
The only fringe benefit with significant and high correlations was free laundry services (r=0.7). This was followed by fringe benefits with moderate relationship and these were insurance coverage (r=0.626), birthday

Discussions
Prior studies provided little information regarding the Malaysian hotel executives' fringe benefits. In order to gain more information and to generalize the findings, a quantitative study focusing on three-, four-and five-star hotels in Langkawi was conducted. This study discovered that the allocation of fringe benefits varies between hotels and it correlates with the hotel's star rating. Both hypotheses were supported by the findings.

Fringe benefit allocation is not the same for various categories of hotels
Except for medical coverage, paid sick leave and paid annual leave, there is a difference in the allocation of fringe benefits to executives in three-, four-and five-star hotels. Out of 22 fringe benefits, the differences of 13 benefits were significant while another nine were not significant. The fringe benefits with insignificant differences were medical coverage, paid sick leave, paid annual leave, holiday entitlement, bonus schemes, clothing provision, free meals, free accommodation and company car.
All executives in this study were entitled for medical coverage, paid sick leave and paid annual leave regardless of the hotel's star rating. This is consistent with the requirement of the Employment Act 1955 which stated that an employee shall be entitled to paid sick leave, after examination at the expense of the employer by a registered medical practitioner appointed by the employer, or by any other registered medical practitioner, or by a medical officer (Government of Malaysia, 1955). When no hospitalisation is necessary, employees are entitled for 14 days yearly sick leave for those who have been employed for less than two years, 18 days yearly sick leave for those who have been employed for more than two years but less than five years, and 22 days yearly sick leave for those who have been employed for more than five years. When hospitalisation is necessary, employees are entitled for 60 days sick leave.
Included in the Employment Act 1955 is the employees' entitlement to paid annual leave and public holiday on ten gazetted public holidays (Government of Malaysia, 1955). Four of the ten public holidays shall be the national day, the birthday of the Yang di-Pertuan Agung, the birthday of the Yang di-Pertua Negeri or the Federal Territory Day and the Workers' Day. This study discovered that almost all executives are provided with this benefit. Not all executives answered "applied completely" to this item. This is due to the executives' perception that they do not have "real" holidays. Due to their commitment, the executives' thought is always Other fringe benefits provided to most of the executives regardless of hotel categories are free meals, clothing provision, bonus schemes and free accommodation. Consistent with Ahmad et al. (2010b) these benefits are provided to executives. However, Ahmad et al. (2010b) stated that there is some variation between the hotels. While some executives have their free meals in the hotel's Coffee House or Executive Lounge, others take their free meals at the staff's cafeteria together with the non-executives. According to the executives, having their meals at the staff's cafeteria gives them the opportunity to maintain good rapport with their subordinates. In addition, it also ensures that their staff are consistently served with appetising food. Staff discount and staff party are also given to most of the executives. However, the level of these two fringe benefits varies significantly between three-, four-and five-star hotels.
Company cars are provided to only a small number of managerial employee regardless of hotel categories. Other fringe benefits that are provided to a few executives but had significant differences among hotels of various categories are loan schemes, pension schemes, relocation allowance, profit-sharing schemes and manager of the month and year award.
Based on the comparison Johnson's (1980) findings, it is concluded that fringe benefits allocation in this study is similar to the developed countries in several aspects. Firstly, it is discovered that the varieties of fringe benefits given to executives in this study are similar with the fringe benefits given to hotel employees in the developed countries. Secondly, fringe benefits such as free food and free accommodation are commonly given in the hotel industry but not in the general industry. Finally, even in the developed countries, fringe benefits such as loan schemes, company cars, personal accident insurance, relocation allowance and profit sharing schemes are given to only a few of the hotel employees.

Fringe benefits allocation correlates with the hotel's star rating
Consistent with previous literature in the developed countries (Guillet, et al., 2012;Johnson, 1983;Kline & Yu-Chin, 2007), this study discovered that the difference in the level of fringe benefits is positively correlated with the hotel's star rating. Out of 13 fringe benefits, the correlations of 12 fringe benefits are significant. The fringe benefit that did not have significant correlation with the hotel's star rating is free transport to and from work.
Fringe benefits with significant correlation with the hotel's star rating are insurance coverage, loan schemes, pension schemes, relocation allowance, personal accident insurance, free laundry services, staff discounts, sports and social facilities, profit-sharing schemes, manager of the month and year award, birthday celebration and staff party. The higher the hotel's star rating, the higher is the level of these fringe benefits allocated to the executives. In other words, five-star hotels provide the highest level of fringe benefits and this is followed by fourand then, three-star hotels.The findings of this study partially support both hypotheses. They are summarised in Table 5.
The findings that the extent of fringe benefits provided is not the same for the various categories of the hotels, and hotel's star rating positively correlates with the amount of fringe benefits provided is consistent with existing literature. Hotels with higher star rating offer a more attractive compensation package (including higher level of fringe benefits) to attract and retain experienced and efficient employees (Cheng & Brown, 1998;Nankervis, 2000).

Conclusion
This research provides the statistics of fringe benefit allocation to executives in three-, fourand five-star hotels in Langkawi. Included is the comparison of the three categories of hotels and the examination of the correlation between the hotel's star rating and the fringe benefit allocation. The statistics reveals that four-and five-star hotels are adopting a more strategic perspective in their fringe benefit packages while three-star hotels are adopting cost control approach. However, all the hotels abide by the Employment Act 1955.
This study contributes to theory and practice. It contributes to theory by providing information on the allocation of various fringe benefits to the executives and how the allocation differs between three-, four-and five-star hotels. Fringe benefits have been excluded in prior compensation research (Pizzini, 2010). This study is valuable to decision makers by providing information on the level of fringe benefits provided by the competitors and the market. Such information is valuable and may serve as a guide for planning competitive fringe benefit packages. As for the hoteliers and hospitality students, this research is useful for their career planning. The information This study has several limitations. Firstly, the non-parametric test which is less powerful compared to the parametric was used because the data was not normally distributed. Secondly, the response rate of less than half of the research sample is low even though it is acceptable. Thirdly, the percentage of respondents from the four-star hotels is high but it is the opposite for the three-and especially the five-star hotels. If there had been more respondents from the fivestar hotels, the findings that the hotel's star rating positively correlates with the level of the fringe benefits allocated could have been consistent across all the items in the fringe benefit packages. Finally, the findings of this research are limited to Langkawi. A similar research with a bigger sample conducted in Kuala Lumpur might have different findings. Future research with a bigger context of study is beneficial for the Malaysian hotel industry.