当前位置:首页 >> 管理学 >>

Applying the stochastic frontier approach to measure hotel managerial efficiency in Taiwan


ARTICLE IN PRESS

Tourism Management 28 (2007) 696–702 www.elsevier.com/locate/tourman

Research article

Applying the stochastic frontier approach to measure hotel managerial ef?ciency in Taiwan
Ching-Fu Chen?
Department of Transportation & Communication Management Science, National Cheng Kung University, Tainan 701, Taiwan Received 22 April 2006; accepted 28 April 2006

Abstract This paper analyses the cost ef?ciency of Taiwan’s international tourist hotel sector. A stochastic cost frontier function with three inputs (i.e. labor, food and beverage, and materials) and one output as the total revenue is speci?ed and used to estimate hotel ef?ciency. The results reveal that hotels in Taiwan are on average operating at 80% ef?ciency. In addition, the factor of operation type signi?cantly affects hotel ef?ciency, whereby the ef?ciency of chain hotels is higher than that of independent hotels. r 2006 Published by Elsevier Ltd.
Keywords: International tourist hotel; Managerial ef?ciency; Stochastic cost frontier

1. Introduction The hospitality industry, especially the hotel industry, is encountering a highly competitive environment worldwide. The formulation of a marketing strategy, strengthening hotel operations, and upgrading the quality of service has become essential not only for pro?tability, but also for a hotel’s survival (Hwang and Chang, 2003). All the factors directly or indirectly depend upon a hotel’s management ef?ciency. In addition, due to its characteristic of an oligopolistic market, the level of competition in an accommodation market demands ef?ciency (Barros, 2004; Phillips, 1999). The issue of ef?ciency may need to be addressed and measured for various reasons (Teague and Eilon, 1973). In terms of strategic reasons, ef?ciency measurement can compare the global performance of an organization with competitors or similar ?rms. In terms of tactical reasons, ef?ciency measurement enables the performance control of an organization or sub-units of it. In terms of planning purposes, ef?ciency measurement can compare the relative bene?ts accruing from the use of different inputs or varying proportion of the same inputs.
?Tel.: +886 6 2757575; fax: +886 6 2753882.

E-mail address: chingfu_chen@yahoo.com.tw. 0261-5177/$ - see front matter r 2006 Published by Elsevier Ltd. doi:10.1016/j.tourman.2006.04.023

Despite the importance of ef?ciency, however, there are relatively few studies in the hotel ?eld (Barros, 2004). Ef?ciency of an organization can be de?ned as a comparison between observed and optimal values of its output and input. In terms of an organization’s behavioral goal, ef?ciency is measured by comparing observed and optimum costs, revenue, or whatever the organization is assumed to pursue, subject to the appropriate constraints on quantities and prices (Lovell, 1993). In the context of a hotel’s operation and management, frontier models can identify inef?cient hotels in a sample via the estimation of inef?cient scores and apply benchmarking criteria to reduce costly inef?ciencies (Anderson, Fish, Xia, & Mixhello, 1999; Barros, 2004; Morey and Dittman, 1995). Differences in location and differences in the quality of a product over a variety of dimensions may create imperfectly competitive conditions for some ?rms. Hence, an examination of the internal ef?ciencies in the workings of ?rms does have value in many industries (Anderson, Fok, & Scott, 2000) without exception towards the hotel industry. Firms can be inef?cient from a failure to allocate resources in the most ef?cient manner (i.e. allocative inef?ciency) and from a failure to utilize their resources given their allocation (i.e. technical inef?ciency) (Anderson et al., 2000).

ARTICLE IN PRESS
C.-F. Chen / Tourism Management 28 (2007) 696–702 697

The term X-inef?ciency ?rst de?ned by Leibenstein (1966) is the difference between how a ?rm could potentially utilize its resources versus its actual utilization. The X-inef?ciency losses arise from managerial inef?ciencies that are a function of inadequate motivation dependent on the extent of market competition. The more competitive the markets are, the more possible it is for the ?rms to operate on their ef?cient frontier. Thus, Anderson et al. (2000) argued that obtaining the X-ef?ciency measures for the hotel industry provides estimates of the competitive structure of this industry. Frontiers have been estimated using many different methods in various empirical studies in the literature. For a well-done introduction to the literature, one can refer to Lovell (1993). The two principal methods that have been used are stochastic frontiers approach and data envelopment analysis (DEA), involving econometric methods and mathematical programming, respectively. The approach adopted by most ef?ciency studies in the context of hotel industry belongs to the latter. DEA assumes that there are no random ?uctuations from the ef?cient frontier, i.e. all deviations are considered inef?ciency. Due to no need to assume the functional form, the DEA is easy to apply but tends to over-estimate inef?ciencies (Anderson et al., 1999). On contrast, the stochastic frontier approach have advantages such as well-developed statistical tests to investigate the validity of the model speci?cation, and ability to decompose the the deviations from ef?cient levels between noise and pure inef?ciency (Barros, 2004). In the context of Taiwanese hotel industry, a few studies, e.g. Tsaur (2001), Chiang, Tsai, and Wang (2004) and Hwang and Chang (2003), have adopted DEA to measure hotel ef?ciencies. However, there is still no study on Taiwan’s hotel sector using the stochastic frontier approach. Hence, this current study adopts the stochastic frontier approach to estimate and analyze the ef?ciency of Taiwan’s international tourist hotel sector. The main purposes of this study are threefold. First, we review the applications of DEA and the stochastic frontier approach to hotel ef?ciency in the literature. Second, we specify a stochastic cost frontier model to estimate the inef?ciency of the hotel sector in Taiwan. Finally, we investigate the ef?ciency level of Taiwan’s hotel sector and in turn examine the effects of hotel management factors (i.e. operation type, hotel location, and hotel scale) on ef?ciency. The structure of this paper is as follows. Section 2 describes the background of Taiwan’s hotel industry. Section 3 reviews the existing literature on hotel ef?ciency. Section 4 speci?es the theoretical framework and sets out the data and results. Following that, the ef?ciencies of the sample hotels are measured and discussed in Section 5, and conclusions are summarized in the end. 2. Background of Taiwan’s hotel industry According to the Taiwan Tourism Bureau, the hotel classi?cation system consists of two groups: international

tourist hotels and ordinary tourist hotels. The international hotels are four-star or ?ve-star equivalent hotels, while ordinary tourist hotels are three-star equivalent hotels. In 2002, there were 62 international tourist hotels which could be classi?ed as independent operations and chain operations (including international and domestic) in terms of ownership and management in Taiwan. Chain operations include the types of franchise chain, management contract, and membership. The independent operation hotel means that an investor owns (or leases) and runs the hotel. The room capacity of these hotels is from a maximum of 873 rooms to a minimum of 50 rooms. However, hotels with 200–400 rooms make up over 50% of all hotels. The 62 international tourist hotels provide a total of 18,790 rooms during 2002 (see Table 1). The construction plan of an international tourist hotel must apply to and be approved from Taiwan’s government. To some extent the international tourist hotel sector is under government regulatory circumstances. 3. Literature review Despite frontier ef?ciency research studies on other industries such as the banking industry and transport industry being common, the analysis of hotel ef?ciency shows up relatively little in the literature. The classical approaches to measure hotel performance are non-frontier models such as the uses of ratio-analysis (Wijeysinghe, 1993), aggregate indices of market performance (Wassenaar & Stafford, 1991), revenue performance (Baker & Riley, 1994), and yield management (Donaghy, McMaton, & McDowell, 1995). In terms of the frontier approach, however, two primary categories can be identi?ed. They are the DEA approach and stochastic frontier approach. The former is adopted by large by most studies. Morey and Ditmam (1995) employed DEA with their input–output data to analyze the ef?ciency of 54 hotels in the United States The input factors used include room division expenditure, energy costs, salaries, non-salary expenses for property, salaries, and related expenses for variable advertising, non-salary expenses for variable advertising, ?xed market expenditures, payroll and related
Table 1 International tourist hotels in Taiwan, 1995–2002 Year 1995 1996 1997 1998 1999 2000 2001 2002 Number of hotels 53 53 54 53 56 56 58 62 Number of rooms 16,711 16,964 16,845 16,558 17,403 17,057 17,815 18,790

Sources: Tourism Statistics, Bureau of Tourism, Taiwan (2003).

ARTICLE IN PRESS
698 C.-F. Chen / Tourism Management 28 (2007) 696–702

expenses for administrative work, and non-salary expenses for administrative work. The outputs used are total revenue, level of service delivered, market share, and the rate of growth. The research found that managers were operating at 89% ef?ciency, and the lodging market appears to be operating ef?ciently. Using the DEA approach, Anderson et al. (2000) analyzed the overall, allocative, technical, pure technical, and scale ef?ciency levels of 48 hotels in the United States. The ?ve inputs used are full-time equivalent employees, the number of rooms, total gaming related expenses, total food and beverage expenses, and other expenses. One output use is the total revenue generated from rooms, gaming, food and beverage, and other revenues. Their results reveal that the hotel industry was operating inef?ciently with a mean overall ef?ciency measure of 42%. Hu and Cai (2004) applied DEA to calculate the labor productivity of sampled hotels in the State of California. In addition, the internal and external determinants of labor productivity are examined in the study. Sigala (2004) developed the stepwise DEA to measure and benchmark the productivity of the UK’s three-star hotel sector. In light of Taiwan’s hotel industry, Tsaur (2000) adopted DEA to analyze the ef?ciency of 53 international tourist hotels there. The inputs used are total operating expenses, the number of employees, the number of guest rooms, the total ?oor space of the catering division, the number of employees in the room division, the number of employees in the catering division, and catering cost. The outputs used are total operating revenues, the number of rooms occupied, average daily rate, the average production value per employee in the catering division, total operating revenues of the room division, and total operating revenues of the catering division. The ef?ciencies of room division and catering division are measured respectively. The results of ef?ciencies for room division and catering division are 71.3% and 89.1%, respectively. Hwang and Chang (2003) used DEA and the Malmquist productivity index to measure the managerial performance of 45 hotels in 1998 in Taiwan and their ef?ciency change from 1994 to 1998. The study shows that the managerial ef?ciency of international tourist hotels in Taiwan is related to the level of internationalization for the hotels. Chiang et al. (2004) also adopted DEA with four input and three outputs to analyze the ef?ciency of 25 four-star or ?ve-star hotels in Taipei. The four inputs used are the number of hotel rooms, the capacity of food and beverage, the number of employees, and the total costs of the hotel. The three outputs used are the yielding index, the revenue of food and beverage, and miscellaneous revenue. The ef?ciencies are measured and compared based upon three types of management i.e. franchise licensed, internationally managed, and independently owned and operated. The results reveal that not all of Taipei’s franchised- or internationally-managed hotels perform more ef?ciently than the independent ones.

In contrast to DEA, however, only two studies use the stochastic frontier approach to analyze the ef?ciency in the hotel industry according to the author’s best knowledge. Anderson et al. (1998) used a classical stochastic frontier model to measure the ef?ciency of 48 hotels in the United States The independent variables of the cost frontier function were the number of full-time equivalent employees, the number of rooms, total gaming related expenses, total food and beverage expenses, and other expenses. They found the hotel industry to be performing relatively ef?ciently, with ef?ciency measures above 90%. Barros (2004) used a stochastic cost frontier to analyze the technical ef?ciency of a Portuguese state-owned hotel chain in order to investigate the chain’s performance. The Cobb–Douglas cost functional form was speci?ed with three input prices (price of labor, price of capital, and price of food) and two outputs (sales and nights occupied). The same data was used to DEA application by Barros (2005). Table 2 summarizes the previous studies on hotel frontier ef?ciency. 4. The stochastic frontier approach The concept of frontier is the main focus of the methods for measuring ef?ciency that have been proposed over the last decade. Ef?cient units are those operating on the cost or production frontier, while inef?cient ones operate either below the frontier (in the case of the production frontier) or above the frontier (in the case of the cost frontier). In the empirical microeconomic literature, there are two groups of methods for estimating frontier functions and thereby measuring ef?ciency: deterministic methods, such as DEA and stochastic frontiers. DEA involves the use of linear programming ?rst introduced by Charnes, Cooper, and Rhodes (1978), whereas stochastic frontiers involve the use of econometric methods ?rst taken up by Aiger, Lovell, and Schmidt (1977) and Meeusen and van den Broeck (1977) simultaneously. The basic stochastic model of the frontier cost (or production) function assumes that any deviation of the observed cost (or production) from the theoretical microeconomic cost (or production) function is caused by purely random disturbances and inef?ciency. The deviation is represented as the composite error term in the stochastic frontier model. The purely random component captures the effect of variables that are beyond the control of the production unit being analyzed (weather, bad luck, etc.). Therefore, a main advantage of the stochastic frontier approach over DEA is that it isolates the in?uence of factors other than inef?cient behavior, thus correcting the possible upward bias of inef?ciency from the deterministic methods. The costs of a company depend on the output vector (y), the price of the input (w), the level of cost ef?ciency (u), and a set of random factors (v). Thus, the cost frontier function is expressed as C ? C?y; w; u; v? ? C?y; w? exp?u ? v?. (1)

ARTICLE IN PRESS
C.-F. Chen / Tourism Management 28 (2007) 696–702 Table 2 Summary of studies on hotel frontier ef?ciency Study Morey and Dittman (1995) Method DEA Units 54 USA hotels Inputs Salaries for major room activities; other room-related expenses; energy cost; salaries for property, operation, and maintenance (POM); other POM; salaries for variable advertising; other variable advertising expenses; ?xed advertising expenses; salaries for administrative and general; other administrative and general expenses The number of full-time equivalent employees; the number of rooms, total gaming related expenses; total food and beverage expenses; other expenses The number of full-time equivalent employees; the number of rooms, total gaming related expenses; total food and beverage expenses; other expenses Total operating expenses; the number of employees; the number of guest rooms; the total ?oor space of the catering division; the number of employees in the room division, the number of employees in the catering division; catering cost Number of full time employees; number of guest rooms; total area of meal department; operating expenses Hotel rooms; food and beverage capacity; number of employees; total cost Number of employees; amount of capital; food and beverage expenses Number of full-time equivalent employees; cost of labor; number of rooms; the surface area of pousada; book value of the premises; operational costs; external costs Outputs Total room revenue; facilitiessatisfaction index; servicessatisfaction index 699

Anderson, Fish, Xia, and Michello (1999)

Stochastic translog production function DEA

48 USA hotels

Total revenue

Anderson, Fok, and Scott (2000)

48 USA hotels

Total revenue

Tsaur (2000)

DEA

53 Taiwan hotels

Hwang and Chang (2003)

DEA

45 Taiwan hotels

Total operating revenues; the number of rooms occupied; average daily rate; the average production value per employee in the catering division; total operating revenues of the room division; total operating revenues of the catering division Room revenues; food nad beverage revenue; other revenues

Chiang, Tsai, and Wang (2004)

DEA

25 Taipei hotels

Yielding index; food and beverage revenue; miscellaneous revenue Operational cost

Barros (2004)

Barros (2005)

Stochastic Cobb–Douglas cost frontier DEA

43 Portuguese hotels 43 Portuguese hotels

Sales; number of guests; aggregated number of nights spent

The composite error term (u+v) is composed of two parts. The ?rst is u, is a one-sided term re?ecting technical inef?ciency, which in the case of the cost frontier is nonnegative and in the case of production frontier is nonpositive. The popular distributional forms for the technical inef?ciency effects are the half-normal, the exponential, and the truncated-normal distributions (Coelli, Prasada, & Battese, 1998; Kumbhakar and Lovell, 2003). The second is v, a two-sided component capturing random shocks and statistical noise, and it is assumed to be distributed as a two-sided normal with zero mean and variance. The frontier function can be estimated by the maximum

likelihood method, as the inef?ciency is estimated from the residuals of the regression. Following Battese and Coelli (1995), the variance terms are parameterized by replacing su2 and sv2 with s ? ?s2 ? s2 ?1=2 ; u v l ? su =sv ;  g ? s 2 s2 . u

The individual estimation of inef?ciency can be obtained using the distribution of the inef?ciency term conditioned to the estimation of the composite error term (Jondrow, Lovell, Materov, & Schmidt, 1982). The cost ef?ciency (denoted as CE) can be de?ned as the ratio of the minimum

ARTICLE IN PRESS
700 C.-F. Chen / Tourism Management 28 (2007) 696–702

feasible cost if the company is ef?cient and the observed costs are CE ? C min C?y; w? exp?v? ? exp??u?. ? C?y; w? exp?u ? v? C (2)

The likelihood ratio statistics are applied to test whether or not the estimate coef?cients are signi?cantly different from zero. 6. Empirical results The maximum likelihood techniques are employed to the estimates of the variable coef?cients and the parameters of the two error components. Table 4 summarizes the estimation results obtained for the stochastic frontier. Except for the occupancy rate (OR) variable, the coef?cients of all variables have the expected signs and are very signi?cant at the 1% level. The ratio of the variability for u and v can be used to measure a hotel’s relative inef?ciency, where l ? su/sv and g ? s2u/s2, and is a measure of the amount of variation stemming from inef?ciency relative to noise for the sample. The values of l and g (i.e. 6.0663 and 0.9735, respectively) reveal that inef?ciency plays an important role in the composite error term and postulate the choice of the stochastic frontier approach in the present study. The cost elasticity with respect to total output is the estimated coef?cient of b3, i.e. 0.9295. The returns to scale of the international tourist hotel sector can be obtained from 1/b3, i.e. 1.0758. This reveals that the operations of the hotel sector are slightly under the situation of slightly increasing returns to scale.
Table 4 Parameter estimates of the Cobb–Douglas cost frontier function Variable label Constant ln(wl/wo) ln(wc/wo) ln TR OR ln VFB s2v s2u s2 ? s2v+s2u l g ? s2u /s2 Log-likelihood Observations Parameter estimate ?4.1916 0.3121 0.5348 0.9295 ?0.4262 ?0.5558 0.00293 0.10785 0.11078 6.0663 0.9735 14.5588 55 t-ratio ?4.250*** 5.238*** 25.915*** 15.433*** ?0.812 ?7.430***

The measure of CE has a value between zero and one. 5. Data and empirical model The empirical data in this study were obtained from the annual report of international tourist hotels published by the Taiwan Tourism Bureau in 2002. Excluding seven hotels due to incomplete data, the operating statistics of 55 international tourist hotels are used to formulate the cost frontier function and estimate the inef?ciency. Among the sample hotels, 33 hotels are located in a metropolitan city and the rest are in a non-metropolitan area. Thirty hotels belong to the group of chain operations while 25 hotels belong to independent operations. The study speci?es a stochastic generalized Cobb–Douglas cost frontier function with three input prices, i.e. price of labor (wl), price of food and beverage (wc), and price of materials (wo), with one output as the total revenue of hotel (TR) and two control variables, i.e. room occupancy rate (OR) and the production value of unit catering space (VFR). The total operating costs (TC), including labor cost, fuel and energy, materials, and external services, are taken as the dependent variable in Eq. (3). The variables are de?ned and characterized in Table 3. Since a cost frontier must be linearly homogeneous in input prices (Kumbhakar and Lovell, 2003), the empirical model with the half-normal inef?ciency assumption is as follows: ln?TC=wo? ? b0 ? b1 ln?wl=wo? ? b2 ln?wc=wo? ? b3 ln TR ? b4 OR ? b5 ln VFR ? ?u ? v?. ?3? The distributional assumptions of error term are as follows: (i) vi$iid N(0, s2v). (ii) ui$iid N+(0, s2u). (iii) vi and ui are distributed independently of each other, and, of the regressors.
Table 3 Main descriptive statistics of variables used in the study (year 2002) Variable TC wl wc wo TR OR VFB Description Total cost (106 NT) Price of labor (103 NT) Price of F&B (103 NT) Price of materials (106 NT) Total revenue (106 NT) Occupancy rate (%) Value produced per F&B space (103 NT) Mean 466.20 462.16 96.94 1.547 568.58 0.594 249.037 Max 1897.50 838.80 232.66 4.212 2636.11 0.803 646.736 Min 14.88 62.96 2.77 0.146 16.96 0.119 3.090 S.D. 450.19 162.96 56.85 0.936 582.33 0.153 153.755

***Shows signi?cance at 1% level.

ARTICLE IN PRESS
C.-F. Chen / Tourism Management 28 (2007) 696–702 Table 5 The estimated ef?ciency of international tourist hotels in Taiwan No. H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 H13 H14 H15 H16 H17 H18 H19 H20 H21 H22 H23 H24 H25 H26 H27 H28 Hotel Grand Hyatt Grand Formosa Regent Howard Plaza Hotel Sheraton Far Eastern Asia World Hotel Grand Hotel Taipei The Sherwood Taipei Ambassador Hotel Hilton Brother Hotel Magnolia Hotel Royal Hotel The Landis Taipei Rebar Holiday Inn Crowne Imperial Hotel The Westin Taipei Golden China Hotel Santos Gloria Hotel Fortuna RiverviewHotel United Hotel Emperor Hotel Grand Hi-Lai Hotel Formosa Regent Kaohsiung Howard Kaohsiung Linden Hotel Kaohsiung Ef?ciency 0.8777 0.9669 0.8214 0.6590 0.6722 0.9270 0.8259 0.9262 0.9173 0.7917 0.7834 0.7404 0.7256 0.8190 0.7600 0.3711 0.7404 0.9064 0.7955 0.8316 0.6343 0.8989 0.8969 0.6576 0.8370 0.7693 0.9156 0.6603 0.8029 0.9771 0.3445 0.1424 No. H29 H30 H31 H32 H33 H34 H35 H36 H37 H38 H39 H40 H41 H42 H43 H44 H45 H46 H47 H48 H49 H50 H51 H52 H53 H54 H55 Hotel Ambassador Kaohsiung Kingdom Hotel Hotel Holiday Garden Royal Hotel Evergreen Laurel Formosa Regent Taichung Hotel National Howard Taichung Plaza International Park Hotel Parkview Hotel Marshal Hotel China Trust Hotel Hualien Hotel Astar Howard Kenting Caesar Park Hotel Royal Chihpen Resort Hotel Grand Formosa Taroko Grand Hotel KH China Yangmingshan Hotel Ta Shee resort Hotel Tainan Hotel Hotel Royal Hsinchu Holiday Inn Toayuan Hotel NanHua The Hibicus Resort Ambassador ShinJu Ef?ciency 0.7826 0.6890 0.9229 0.3445 0.9425 0.9396 0.8107 0.8947 0.8801 0.5569 0.8392 0.7671 0.8583 0.9523 0.9234 0.7762 0.8532 0.8498 0.6849 0.9771 0.8667 0.9235 0.8873 0.8995 0.3993 0.8822 0.9280 701

Mean ef?ciency Highest ef?ciency Lowest ef?ciency Standard deviation

The ef?ciency level of each hotel can be obtained from Eq. (2) with the estimated inef?ciency u. Table 5 shows the results of hotel cost ef?ciency. The average ef?ciency is 80.30% and indicates that almost 20% costs can be reduced without decreasing output if the hotel can operate ef?ciently. The maximum hotel ef?ciency score is 97.72% while the minimum ef?ciency score is 34.46%. Thirty three out of 55, i.e. 60%, hotels are operating with an ef?ciency higher than the average ef?ciency. Compared to the ?nding elsewhere in the same industry, these ef?ciency scores are lower than the 90% in the United States (Anderson et. al., 1999), but much higher than the 21.6% in Portugal (Barros, 2004). The low ef?ciency in the latter case is attributed to its characteristic of a state-owned enterprise. To investigate the effects of the factors of management in interest, the one-way ANOVA analyses are conducted. Table 6 presents the results of ef?ciency difference by three factors. Only the factor of operation type (i.e. chain operation vs. independent operation) reveals the signi?cant difference on an ef?ciency score at the 10% signi?cance level. In addition, the ef?ciency of chain operation (mean ? 0.8345) is higher than that of independent operation (mean ? 0.7651). Hwang and Chang (2003)

Table 6 One-way ANOVA results for the affecting factors on ef?ciency Mean ef?ciency Operation Type Chain (n ? 30) Independent (n ? 25) Location Metropolitan (n ? 33) Non-metropolitan (n ? 22) Scale Small:o300 rms (n ? 31) Large:4300 rms (n ? 24) F value 3.382 0.8345 0.7651 1.782 0.7822 0.8341 0.129 0.8090 0.7950 0.721 0.188 Sig. 0.072

provided a similar ?nding. Note that the chain operation in this study includes both international chain and domestic chain while Hwang and Chang (2003) only specify the international chain. Despite being insigni?cant, the result of the effect of hotel location on ef?ciency is still informative. The ef?ciency of hotels located in the metropolitan area (mean ? 0.7822) are lower than in the non-metropolitan area (mean ? 0.8341). This ?nding is

ARTICLE IN PRESS
702 C.-F. Chen / Tourism Management 28 (2007) 696–702 Aigner, A., Lovell, C. A. K., & Schmidt, P. (1977). Formulation and estimation of stochastic frontier production function models. Journal of Econometrics, 86, 21–37. Baker, M., & Riley, M. (1994). New perspectives on productivity in hotel: some advances and new directions. International Journal of Hospitality Management,, 13(4), 297–311. Barros, C. P. (2004). A stochastic cost frontier in the Portuguese Hotel Industry. Tourism Economics, 10, 177–192. Barros, C. P. (2005). Measuring ef?ciency in the hotel sector. Annals of Tourism Research, 32(2), 456–477. Battese, G. E., & Coelli, T. J. (1995). A model of technical inef?ciency effects in a stochastic frontier production function for panel data. Empirical Economics, 20, 325–332. Charnes, A. C., Cooper, W. W., & Rhodes, E. (1978). Measuring the ef?ciency of decision making units. European Journal of Operational Research, 2(6), 429–444. Chiang, W., Tsai, M., & Wang, L. S. (2004). A DEA evaluation of Taipei hotels. Annals of Tourism Research, 31, 712–715. Coelli, T., Prasada, R., & Battese, G. (1998). An introduction to ef?ciency and productivity analysis. Dordrecht: Kluwer Academics Press. Donaghy, K., McMahon, U., & McDowell, D. (1995). Yield management: An overview. International Journal of Hospitality Management, 14(2), 1339–1350. Hu, B. A., & Cai, L. A. (2004). Hotel labor productivity assessment: A data envelopment analysis. Journal of Travel and Tourism Marketing, 16, 27–38. Hwang, S., & Chang, T. (2003). Using data envelopment analysis to measure hotel managerial ef?ciency change in Taiwan. Tourism Management, 24, 357–369. Jondrow, J., Lovell, C. A. K., Materov, I. S., & Schmidt, P. (1982). On the estimation of technical inef?ciency in the stochastic frontier models. Journal of Econometrics, 19, 233–238. Kumbhakar, S. C., & Lovell, C. A. K. (2003). Stochastic Frontier Analysis. Cambridge: Cambridge University Press. Leibenstein, H. (1966). Allocative ef?ciency vs ‘‘X-ef?ciency’’. American Economic Review, 56(3), 392–414. Lovell, C.A.K. (1993). Production frontier and productivity ef?ciency. In L. Fried, et al. (Ed.), The measurement of productive ef?ciency: Techniques and applications. New York: Oxford University Press. Meeusen, W., & van den Broeck, J. (1977). Ef?ciency estimation from Cobb–Douglas production function with composed error. International Economic Review, 18, 435–444. Morey, R., & Dittman, D. (1995). Evaluating a hotel GM’s performance: A case study in benchmarking. Cornel Hotel Restaurant and Administration Quarterly, 36(5), 30–35. Phillips, P. A. (1999). Performance measurement systems and hotels: A new conceptual framework. International Journal of Hospitality Management, 18(2), 171–182. Sigala, M. (2004). Using data envelopment analysis for measuring and benchmarking productivity in the hotel sector. Journal of Travel and Tourism Marketing, 16, 39–60. Teague, J., & Eilon, S. (1973). Productivity measurement: A brief survey. London: Imperial College of Science and Technology Chapman and Hall. Tsaur, S. (2001). The operating ef?ciency of international tourist hotels in Taiwan. Asia Paci?c Journal of Tourism Research, 6, 73–81. Wassenaar, K., & Stafford, E. R. (1991). The Lodging Index: An Economic Indicator for the Hotel Motel Industry. Journal of Travel Research, 30, 18–21. Wijeysinghe, B. S. (1993). Breakeven Occupancy for a Hotel Operation. Management Accounting, 71, 32–33.

also in line with Hwang and Chang (2003). The former mainly is made up of business hotels which primarily offer accommodations, while the later mainly targets the leisure market. The perspectives of product mix and sources of customers reveal that the hotels targeting the leisure market have more advantages than business hotels in effectively utilizing their operation resources and turning into higher managerial ef?ciency. Finally, similar to Hwang and Chang (2003), no difference in ef?ciency can be found between the large-scale hotels and the small-scale hotels. 7. Concluding remarks Facing strong growing competition, the ef?ciency of hotel operations and management plays a crucial role to determine a hotel’s pro?tability and even its survival. Ef?ciency measures can provide hotel managers with benchmarking information and further insight on the improvement of resources’ utilization. DEA and stochastic frontier approaches are the two main methods to estimate the ef?ciency in terms of the frontier concept based on production theory. Although DEA has been adopted by most hotel ef?ciency studies, instead of DEA this study chooses the stochastic frontier approach due to its advantage on the capability to isolate the in?uence of factors other than inef?cient behavior. This paper analyzes the cost ef?ciencies of 55 Taiwanese international tourist hotels in 2002. A stochastic cost frontier function with three inputs (i.e. labor, food and beverage, and materials) and one output (as total revenue) is speci?ed and used to estimate hotel inef?ciency. The results reveal that the hotels are on average operating at 80% ef?ciency and the market is competitive in general. In addition, management type appears to signi?cantly affect hotel ef?ciency. The ef?ciency of chain hotels is higher than that of independent hotels, which reveals that the management system or brand equity chain hotels can bring a positive impact on a hotel’s ef?ciency on average. However, no signi?cant evidence can be found that ef?ciency is affected by both hotel location and scale. Inef?cient hotels can look for their ef?ciency improvement strategies based upon the benchmark of those ef?cient hotels. References
Anderson, R. I., Fish, M., Xia, Y., & Mixhello, F. (1999). Measuring ef?ciency in the hotel industry: A stochastic approach. International journal of hospitality Management, 18(1), 45–47. Anderson, R. I., Fok, R., & amd Scott, J. (2000). Hotel industry ef?ciency: An advanced linear programming examination. American Business Review, 18(1), 40–48.


赞助商链接
相关文章:
更多相关标签: