Introduction

The positive evolution of the demand for hotel services has boosted the role of hotel corporations in the Spanish hotel offer to the detriment of independent hotels. In fact, according to the data of Hotelmarket for 2015, the vast majority of the Spanish hotel offer is in the hands of hotel corporations in a percentage close to 80%. However, this type of large companies has also been affected by the recent economic crisis that began in 2007, although its impact has not been homogeneous. While some hotel groups have had to declare bankruptcy or carry out refinancing of their liabilities, as is the case of Hoteles Foxá, the Urvasco group, and High-tech Hotels, others have boosted their plans for international expansion, especially in Latin America, as is the case of Meliá Hoteles, NH hotels, and Room Mate.

This change in the general structure of the sector has consequences for different interest groups given that, as a general rule, hotel groups have a greater dimension and, consequently, their decisions and actions have effects on a wider community of agents. In this line, it should be noted that tourism, where the hotel sector is its main component, represented 10.2% of Spanish GDP in 2010 and increased to 11.1% in 2015, according to data published by the National Statistics Institute.

The consequences of the prominence achieved by hotel corporations in the Spanish offer reach other more particular factors, such as the management of their performance. This is due to the fact that, generally speaking, in this type of companies it is their asset management model and their possibility of generating and taking advantage of synergies that-for example-are translated into economies of scale, which have an impact on their profitability. Furthermore, it must be considered that the very survival of the hotel is affected by the very management given by these large companies, which is not without complexity and requires an in-depth study of the factors that may influence it (^{Li and Sun, 2012}). Thus, the focus of this work is on analyzing business performance, measured through economic profitability because this indicator relates business results with investment and is the measure most used in previous studies (^{Sainaghi et al., 2017}; ^{Lado-Sestayo et al., 2016a}).

The competitive environment is also not alien to the structural change caused by the greater presence of hotel groups (^{Becerra et al., 2013}; ^{Sami and Mohamed, 2014}). This may be related to the emergence of integration processes that seek to maintain or strengthen the competitive position of hotel establishments in a market that, as is the case of Spain, currently has an offer dominated by this type of companies that generate a high volume of employment (^{Vivel-Búa et al., 2016}).

In short, this scenario justifies the relevance of focusing a study on this type of hotel companies. Thus, the objective of this empirical study is to analyze the performance, measured through economic profitability, of hotels integrated in hotel corporations within the Spanish market during the 2005-2011 period. In particular, it evaluates the factors that may influence this economic indicator, considering both those intrinsically related to the hotel establishment and others more closely related to the tourist spot where it is located. Thus, the consideration of these two dimensions (hotel and location) constitutes one of the main contributions of this research, as most previous studies focus on variables at a company level (^{Ivanov and Zhechev, 2012}). At the same time, given the international importance of Spain as a tourist market, this country is the focus of outstanding analysis in order to obtain results that can be extrapolated to other markets. According to the World Tourism Organization (WTO), it should be noted that Spain is among the five most important international destinations in both tourist arrivals and income generated by tourism. In addition, it is the first country in the world in tourism competitiveness according to the ranking of the World Economic Forum for 2015.

The work is organized in five sections. After this first introductory section, the second section exposes the theoretical framework to which this research subscribes, showing the related literature and the hypothesis approach. The third section identifies the database used and explains the empirical study carried out. This work ends with a fourth section presenting the conclusions and, subsequently, the compilation of bibliographical references in the fifth section.

Theoretical framework

Despite the importance of hotel corporations in the supply of tourism services, there are few studies focused on evaluating the determinants of their performance. Nevertheless, the previous literature does recognize the existence of multiple differences between hotels integrated in these business groups and independent hotels (^{Gursoy and Swanger, 2007}; ^{Botti et al., 2009}). For example, ^{Sami and Mohamed (2014)} find that hotel chains perform better than other types of hotel companies, although they also show a negative relationship between their size and performance. In fact, focusing on the nature of the determinants of hotel performance, it should be noted that the characteristics of the hotel have been the main factors analyzed in the previous literature (^{Ivanov and Zhechev, 2012}).

From the point of view of business models, there are differences between independent hotels and those integrated into a hotel corporation (^{Yeung and Lau, 2005}; ^{Damonte et al., 1997}). This can be related to the knowledge economies obtained by hotels already integrated into the corporation, and even to the increase in performance obtained by increasing the market share resulting from affiliation (^{Yeap, 2011}; ^{Sinclair and Stabler, 1997}; ^{Zhao, 1994}). In this respect, previous studies have found that the higher market share may have strategic consequences by increasing the ability of the hotel corporation to influence price in the interrelations with its competitors (^{Yeap, 2011}). Consequently, the market share will be the first factor considered in this study and the following hypothesis will be posed:

One of the variables analyzed as a determinant of performance is the number of stars, which presents a positive relationship with price and consequently increases profitability (^{Lado-Sestayo et al., 2017}; ^{Becerra et al., 2013}). ^{Becerra et al. (2013)} find that hotels affiliated with hotel corporations have higher prices and lower discounts, although these authors affirm that this horizontal differentiation is less effective than the differentiation in the level of quality (vertical). Consequently, because the competitive environment is exogenous, hotels must design strategies to reduce competition and increase their performance, for example, by enhancing their differentiation (^{Graff, 2011}). Thus, the number of stars will be the second factor considered and according to previous literature the following hypothesis is explored:

A very popular horizontal differentiation strategy is the affiliation of a hotel to hotel corporations. According to previous studies, this can lead to a reduction in the probability of insolvency and favor greater economies of scale (^{Baum and Ingram, 1998}; ^{Sinclair and Stabler, 1997}). Moreover, this integration can generate externalities (^{Lee and Jang, 2015}; ^{Tsang and Yip 2009}). In fact, these may refer to knowledge sharing, or the concentration of employees with a higher degree of specialization and productivity (^{Glaeser and Resseger, 2010}; ^{Halling and Marnburg, 2008}; ^{Desrochers, 2001}). At the same time, these externalities may create an environment conducive to the sharing of infrastructure or social capital between hotels within the same corporation (^{Kalnins and Lafontaine, 2004}). However, it should be noted that, according to recent previous studies, this positive effect requires incurring costs derived from the lower efficiency associated with larger size (^{Sami and Mohamed, 2014}). Likewise, hotel corporations have higher fixed costs, which, although they imply economies of scale in periods of growth, can also, in periods of low demand, cause greater difficulties and compromise the survival of the hotel (^{Lado-Sestayo et al., 2016b}). Consequently, the size of the hotel, as a proxy for the possible existence of economies of scale, is the third factor considered in this work and, according to ^{Lee and Jang (2015)}, the following hypothesis is considered:

Within the hotel factors that determine profitability, previous studies have considered a set of control variables. Thus, the level of debt due to pressure on the strategic decisions of the company affects performance (^{Liu & Hung, 2006}; ^{Vivel-Búa et al., 2018}). Another variable considered is the level of liquidity, because a deficit or excess can have significant consequences on hotel management (^{Borde, 1998}; ^{Jensen, 1986}). This relationship has been examined in numerous studies, although the sign of its effect is not conclusive (^{Kim et al., 2012}; ^{Chen, 2013}). Therefore, both variables (indebtedness and liquidity) together with an indicator of the ability of a company to meet its short-term payments, measured through working capital, will be considered in this study.

External factors can also significantly affect performance, according to the literature. In fact, ^{Sami and Mohamed (2014}) find a notable effect of such variables on performance, and ^{Sainaghi (2011}, ²⁰¹⁰⁾ stresses the need to consider the geographic location of the hotel establishment in assessing its performance in a sample of independent hotels. Similarly, ^{Yang et al. (2015)} show that it is necessary to incorporate characteristics related to the destination where the hotel is located to analyze its performance. Thus, the level of demand has been shown to have a positive impact on performance (^{Lado-Sestayo et al., 2016}; ^{Lado-Sestayo et al., 2017}). The level of demand will be the fourth factor considered in this study based on the following hypothesis:

In order to reflect the heterogeneity in hotel demand between tourist destinations, due to the fact that in Spain sun and beach destinations with high seasonality coexist with urban, cultural, and rural destinations, the level of seasonality has been considered as a study variable (^{Ridderstaat et al., 2014}). According to Vivel Búa et al. (2017), seasonality has a negative impact on performance due to the fact that the sector has high fixed costs. Therefore, in contexts of high seasonality, managers must adapt resources to market demands (^{Karamustafa and Ulama, 2010}). This is particularly difficult in independent hotels and, consequently, affiliated hotels may have an advantage in sharing infrastructure. In addition, the temporary concentration of demand may result in an increase in prices that, together with the better development in these contexts of affiliated hotels due to their capacity to reduce resources and have a better income management, would have a positive effect on their performance (^{Vives et al., 2018}). Thus, the following hypothesis is considered:

Finally, the last factor to consider is the competitive environment. Thus, the existence of a concentrated competitive environment, where collusive practices can be carried out, could increase profitability (^{Lee, 2015}). According to these results, hotel corporations competing in multiple markets could present greater incentives for collusion and, therefore, better performance (^{Mauri, 2016}). Consequently, market structure will be the fifth factor considered in this study. In this line, the affiliation processes could contribute to the establishment of entry barriers against new competitors, which could favor the increase of the price level and, consequently, of the performance of the affiliated hotels (^{Lado-Sestayo et al., 2016a}; ^{Lee, 2015}; ^{Conlin and Kadiyali, 2006}; ^{Zhao, 1994}; ^{Suzuki, 2013}).

According to the postulates of the Structure-Conduct-Result (SCP), the existence of collusion among the installed companies would justify the relationship between market concentration and performance (^{Cowling and Waterson, 1976}; ^{Davies, 1999}). This paradigm establishes an analytical framework on which business results depend on the behavior of companies, which in turn depend on the structure of the market. Thus, the structure of the market characterized by the degree of concentration of supply, the degree of concentration of demand, the level of product differentiation, and entry barriers determines the competitive intensity. Within this framework, companies and consumers decide on their strategies, which determines the business outcome.

The Chicago School also finds a positive relationship between market concentration and performance, although this is based on efficiency, which is higher in large companies (^{Demsetz, 1973}; ^{Williamson, 1975}, ¹⁹⁸⁵). Given that this research uses a proxy of economies of scale (company size), it is feasible to validate the SCP approach in line with previous studies including the market structure as a determining factor (^{Lado-Sestayo et al., 2016a}), since the effect of size is collected through the proxy of economies of scale:

The hypotheses presented will allow a better understanding of the determinants of the performance of hotels affiliated to hotel chains, helping to identify the main differences between these and independent hotels.

Empirical analysis

Results analysis

Table 2 presents the results obtained through the econometric estimates made. In particular, model 1 refers to the estimation by OLS, model 2 to the estimation of panel data with random effects, and model 3 to the estimation with fixed effects. According to the statistics calculated to validate these estimates, it can be concluded that the model with fixed effects is preferable to the rest of the estimates (model 3). This is due to the fact that the Fixed Effects F Test confirms that these have statistical significance, just as the Random Effects LM test confirms the significance of random effects. Therefore, both models (fixed effects and random effects) are preferable to the model estimated by OLS. In order to select between the fixed-effects model and the random-effects model, the Hausman test was performed. Its results suggest that the fixed-effects model is preferable to the random-effects model.

The results show that demand and market structure are relevant variables that influence the performance of Spanish hotel groups. Therefore, this stresses the relevance of the characteristics of the tourist spot when evaluating hotel performance.

Focusing on the characteristics of the hotel, the estimates confirm the effect of market share and economies of scale. With respect to the former, this result confirms that the market share does not necessarily have to be increased by sacrificing the profit margin. In relation to economies of scale, these can be individually exploited by each hotel establishment integrated in a hotel group.

No evidence has been found for the quality of the hotel measured by its star rating. Perhaps this is due to the fact that quality not only influences price but also costs. This could mean that other factors apart from the quality considered are the only ones that can have an impact on hotel performance.

Focusing on control variables, the volume of debt has a negative effect on performance. This could relate to the influence that debt can have on strategic decisions and price related tactics, which could be oriented towards generating cash flow rather than improving performance.

For its part, employment has a negative relationship with hotel performance. This could be justified by the associated costs or even by the marginal efficiency associated with the increase in staff.

With regard to cash flow, this has a positive impact on the performance of the hotels integrated in the hotel corporations.

Finally, the method proposed by Billor et al. (2000) has been used to detect the possible existence of outliers with incidence in the estimated parameters. The results confirm the absence of outliers.

Conclusions

This work has studied the determinants of the performance of 1,019 hotels affiliated to hotel corporations in Spain in the 2005-2011 period. To this end, both variables that individually characterize each hotel and variables that characterize the Spanish tourist destination where the hotel is located have been considered. The incorporation of both dimensions, i.e., hotel and tourist spot, means that this work contributes to the low number of studies that include variables external to the hotel in the study of its performance. Likewise, the majority of previous studies have focused on independent hotels despite the fact that notable differences have been found between these and hotels affiliated with hotel corporations.

The results obtained have contributed to the support of the hypotheses regarding the positive effect of market share, economies of scale, level of demand, and market concentration on performance. This means that hotels with a large presence in the market and a larger size have a better performance, which is in line with what was obtained in previous studies. Therefore, those actions aimed at improving market share are a way to improve results. This result confirms the strategic impact of the market share on competitors highlighted in previous studies. Likewise, with regard to the positive impact of size, the existence of economies of scale is confirmed, which is in line with the existence of positive externalities evidenced in previous literature, justifying the greater presence of hotel corporations in the market through the study horizon considered. This could be explained by the fact that the affiliated hotels share advertising and management costs, as well as having greater negotiating power vis-a-vis marketers and particularly vis-à-vis online travel agencies, which concentrate a large part of the hotel offer.

Regarding the hypotheses that have not been verified, no evidence has been obtained that supports the existence of a positive effect of the level of quality or of the variability of demand on performance. This could be due to the horizontal differentiation presented by affiliation, which could compensate for the absence of differentiation in quality via a greater number of stars. Vertical differentiation (greater number of stars) implies costs in line with the greater income it generates. In summary, the divergence in this result with previous research analyzing independent hotels could be found in the fact that the level of quality is an indicator that favors greater demand in independent hotels, although the image of the hotel corporation acts as an indicator in the affiliated hotels. With regard to the absence of the effect of seasonality, the results indicate that the best position of affiliated hotels to adapt resources to market demands according to previous literature is not sufficient to compensate for the effect of fixed costs.

In relation to the control variables used, the results point to the level of indebtedness negatively affecting performance, while liquidity contributes to its increase. Given the lack of previous studies that-from a theoretical point of view-support this relationship in the sector, these results point to the need to deepen the impact of the financial management of the hotel, given that it could play an important role in generating positive results. Among other aspects, this could be explained by the cost structure in the sector, very dependent on fixed costs and by the fact that the product offered is perishable.

The results show that the analysis of the performance of hotel corporations must consider variables of the hotel and tourist destination and that, consequently, the models proposed for independent hotels in previous literature may not be valid for affiliated hotels. In addition, these results present areas of common interest for hotel managers and tourist destinations, for example, for the promotion of tourist destination demand. However, they also point out that when faced with other types of actions-for example, those aimed at reducing seasonality-, affiliated hotels may not have incentives to collaborate with destination managers.

There were some limitations to the development of this work. Despite the effort made to obtain complete information on the hotel corporations to be analyzed, this was limited by the availability of data, conditioning and limiting the study period. Likewise, it was not possible to have information on the tourist spot in the period prior to 2005.