Choice of entry mode, strategic flexibility and performance of international strategy in hotel chains: an approach based on real options

This paper presents an explanatory model for the choice of entry mode to an international market and its implications for the performance of the international strategy in hotel chains, under the dynamic approach of real options theory. The proposed model includes the analysis of some alternative variables to those already investigated in the literature, such as the perception of business uncertainty, international strategic flexibility, and the effect on performance of the international strategy based on the possession and implementation of real options. Our results show that high levels of perceived uncertainty (exogenous and endogenous) and the interaction between the size and international experience of the company are negatively related to levels of international property investment, which allows the company to possess and implement real options conferring sufficient strategic flexibility to adapt its strategy in response to changes, in order to improve organizational performance. We also found that the possession of real options can negatively influence the image and the degree of access to a foreign market if the company exercises the option to leave that market. © 2015 International University College. All rights reserved


Introduction
The object of the present research is to gain more in-depth knowledge of the process of internationalization of companies. We focus on organisations belonging to the tourism sector, and specifically on the Spanish hotel sector, including both the national chains and those international groups that operate in Spain. The international behaviour of the foreign hotel chains follows a similar pattern in all the countries in which they may decide to operate; therefore the characterization of this behaviour should be of interest for any country that wishes to attract investment from organisations of this type. Equally, although the Spanish hotel chains have traditionally concentrated their international investments in the Caribbean region, in recent years the European destinations have been acquiring increasing importance; for this reason it is very important to understand the particularities of the international strategy followed by Spanish hotel chains, in order to attract their investment or if potential strategic alliances with these chains is being considered.
Our interest in companies of this type is due both to the importance of this sector in the Spanish economy -where tourism activity, according to the Instituto de Estudios Turísticos (IET, 2012) represents around 10% of GDP and a total of 281.4 million overnight hotel stays were recorded in 2011 -and to the relative scarcity of studies centred on this topic. Although the decision by a company on the best mode of entry to a new country market has been widely studied in the manufacturing sector (Gatignon and Anderson, 1988;Kogut and Singh, 1988;Hill et al., 1990;Agarwal, 1994;Tse et al., 1997;Arora and Fosfuri, 2000;Pang and Tse, 2000;etc.), the conclusions reached in these studies cannot easily be translated to the services sector in general, nor to the hotel sector in particular; in the latter sector the most recent research demonstrates the differences that are presented versus the manufacturing sector (Agarwal and Ramaswami, 1992;Kundu, 1998a, 1998b.;Ramón, 2002;Quer et al., 2007;Graf, 2009;Martorell et al., 2013). Nevertheless, in our case, the specific objective of the research study is more limited.
Our study presents an explanatory model of the choice of the mode of entry to a foreign market and of the repercussions of this choice on the result of the international strategy of the hotel chains, under the dynamic approach of the theory of real options and giving refined treatment to the topic of uncertainty (Miller, 1993;Werner et al., 1996;Brouthers et al., 2000Brouthers et al., , 2002Li and Rugman, 2007). The applicability of this theory to this field of study is analysed from an integrative conceptual approach, which complements the perspective of real options with concepts derived from other theories (theory of transaction costs, agency theory, theory of behaviour and theory of resources and capacities of the company), in order to explain the modality of entry to foreign markets chosen by companies of the hotel industry (Contractor and Kundu, 1998a;Ramón, 2000Ramón, , 2002Berbel, 2006;Martorell et al., 2013). This model will allow us to determine the significant factors that the directors take into account in managing the strategic fit, faced with changes in the initial conditions of uncertainty, by means of strategic flexibility and real options.
In summary, the objectives set for this study are: 1. To check the applicability of the theory of real options to the processes of internationalization, and to confirm its explanatory capacity in the hotel industry; and, in doing so, to add new factors to the doctrinal body existing on the mode of entry to a foreign market in the hotel industry. To the best of our knowledge, this study pioneers research in which this theory is incorporated into the international strategy in companies of this type. 2. To deal in greater depth with the topic of perceived uncertainty in the business environment and its influence on the choice of the mode of entry, from the perspective of real options. To test empirically, for the first time, the hypotheses put forward by Li and Rugman (2007), which distinguish between two types of uncertainty, exogenous and endogenous. 3. To identify the variables that influence the choice of the particular mode of entry. 4. To understand how the mode of entry can contribute to the creation of real options available to the company. 5. To analyse the influence of the existence of real options on the company's strategic flexibility. 6. To check the ways by which the existence of real options and strategic flexibility can influence the company's international performance.
In short, our research is intended to answer the following questions: -Does international uncertainty influence the mode of entry chosen by a hotel chain? -Do the structural variables of the company influence the way the company expands internationally? -Do the strategy of the company and factors related to its control influence the volume of investment in international markets? -How does the choice of mode of entry to a foreign market influence the real strategic options open to the company? -Is there a relationship between the real options open to the organisation and its strategic flexibility?

Literature review
The concept of real options was developed by Myers in 1977, and refers to the possible application of the theory of options for valuing non-financial goods, and specifically to the investment in real assets that present a component of flexibility, such as investment in research and development or the expansion of manufacturing plants.
Financial options give the holder the right, but not the obligation, to sell or purchase shares at a predetermined price, known as the exercise price, during a period of time also predetermined. Therefore, financial options remain subject to market fluctuations and present an asymmetric distribution of return, derived from the right, but not the obligation, of exercising the option (Amram and Kulatilaka, 1999;Trigeorgis, 1996).
Hence, the approach of real options is nothing more than the extension of the Theory of Financial Options to options in real (nonfinancial) assets that allow the modification of an investment project with the intention of increasing its value, since real options offer a company the opportunity to increase its gains or reduce its losses (Calle and Tamayo, 2009). Myers (1977) stated that non-recoverable investments create real options because they provide discretional opportunities in the future. In addition, Bowman and Hurry (1993) maintain that the resources of an organisation, its capacities and assets, can be seen as a set of real options for future strategic choices.
Although there has been increasing academic interest in recent years in the application of the theory of real options to international strategy (Capel, 1992;Chi and McGuire, 1996;Petersen et al., 2000;Tong et al., 2008;Driouchi and Bennett, 2011;Fisch, 2011), we are faced with a relatively new and ill-defined field that presents great potential for tackling some of the more complex questions in this field, such as analysis of the strategies of the multinational companies in conditions of risk and international uncertainty.
The application of the theory of real options to international strategy is a recognition that uncertainty is associated not only with risks but also with potential opportunities which multinational companies should be able to exploit (Chi and McGuire, 1996;Tong and Reuer, 2007). This uncertainty creates real options that allow them to maintain the necessary flexibility for making the adjustments that will be needed as they decide how to respond to new opportunities or challenges that arise (Kogut and Kulatilaka, 1994). Li (2007) proposed the logic of the theory of real options and its applications to international strategy: to generate improved performance, a multinational company must first evaluate the different types of uncertainty in the chosen market, exogenous and endogenous.
We should make it clear that exogenous uncertainty is that which is not affected by the decisions and actions of a company; it can only be identified with time, like uncertainty regarding political decisions and the macroeconomic environment. Endogenous uncertainty can be reduced by measures taken by the company itself; included in this category are uncertainty regarding the microeconomic environment (such as the needs of consumers and the conditions of competition) and uncertainty at the level of the company (Miller, 1993;Li, 2007).
Then the company creates real options through the construction of a multinational network and the choice of the best time and mode of entry into that market. The value of the real options is materialized when the multinational company exercises those options in response to new information. Therefore, the real options generate value for the multinational company (by increasing the market value and diminishing the risks of the company), when the complete process of generation and execution of options is correctly applied (Capel, 1992;Petersen et al., 2000;Copeland and Tufano, 2004;Tong et al., 2008;Driouchi and Bennett, 2011;Fisch, 2011).

Specification of the model
Our starting point is the model of Contractor and Kundu (1998a); this model analyses the various possible modes of entry previously utilized by the world hotel industry, and the factors that determine the mode selected. In accordance with this, the alternative modalities of entry in external markets consist of the following: franchising, management contracts, leasing contracts, and direct investment (a distinction is also usually made between full and shared ownership, or joint venture). Adapting this model on the basis of the theory of real options, we put forward a new model intended to explain the decisions of a hotel company regarding the mode of entry into an international market, by analysing the direct influence of diverse variables -as was done by Contractor and Kundu (1998a), Ramón (2000Ramón ( , 2002, Berbel (2006), Quer et al. (2007) and Martorell et al. (2013), and relating them to the result obtained. However, the model put forward here goes further in the analysis of alternative variables to those traditionally  investigated by other authors, such as: the perception of the level of uncertainty; international strategic flexibility; and the result of the international strategy as a consequence of the possession and exercising of real options. Therefore, we take into consideration the contribution of the theory of transaction costs (Coase, 1937;Williamson 1975Williamson , 1979Williamson , 1985, which suggests that the level of idiosyncrasy and irreversibility of the investments made in the hotel industry limits the strategic flexibility of this type of organisation in its international Strategy, and we complement it with the perspective of the theory of real options, in order to increase understanding of the decision on the mode of entry to a foreign market.
In fact, our final model responds to the following reasoning: in the face of high levels of uncertainty the company will opt for the modality of entry that gives it valuable real options; that is, the entry mode that offers it sufficient strategic flexibility to be able to adapt its strategy in response to changes in the degree of perceived uncertainty, with the object of maximizing the organizational performance. Figure 1 summarises the model presented.

Justification of the hypotheses
The theory of real options emphasizes that, in situations where uncertainty does not allow the value of an investment opportunity to be predicted with accuracy, the company would respond by making the minimum initial investment that would leave open options of greater investment in the future (Brouthers et al., 2008); in this case, a cooperative mode of entry would be an attractive option. Although most attempts at modelling the real options in the international strategy of companies include uncertainty as an independent variable, these do not differentiate between the ways in which the various types of uncertainty condition the company's mode of entry into the international market. In this respect, Li and Rugman (2007) use modelling and simulation techniques to conclude that the choice of the mode of entry depends both on the magnitude (high or low) of the uncertainty and on its type (exogenous or endogenous). In line with these authors, we put forward the following hypotheses: To answer the question of how the size and international experience of the expanding company affects the mode of entry, we find two differing opinions. A majority line of theoretical argument indicates that the larger and the greater the international experience of the investing company, the greater the probability that it would opt for a direct investment (Erramilli, 1991;Agarwal and Ramaswami, 1992;Contractor and Kundu, 1998a). However, we find other lines of theoretical study that suggests the opposite (Anderson and Gatignon, 1986;Ramón, 2000Ramón, , 2002Berbel, 2006). In the global hotel environment, given that this industry requires a high level of investment that has the effect of encouraging strategic alliances, it is the more experienced companies and those of greater size, whose greater volume of activities similarly generates international experience (Contractor and Kundu, 1998a, 1998bRamón, 2000Ramón, , 2002Berbel, 2006) that have sufficient prestige and resources for forging alliances and finding strategic partners more easily. Therefore, we propose the following hypothesis: Hypothesis 3: The interaction between company size and international experience is associated negatively with the mode of entry.
The perception that the directors of a multinational company have regarding certain variables related to the global strategy of the company in the medium and long term, must also influence the choice of mode of entry. This means that the possible internationalization of a company must be considered from a dynamic perspective. Hence, we propose testing the information obtained regarding the variables traditionally considered in studies on the choice of the mode of entry of the international hotel chains, which respond to two basic strategic objectives -growth and cost reduction, on the one hand, and market differentiation, on the other (Contractor and Kundu, 1998a;Ramón, 2000Ramón, , 2002Berbel, 2006;Martorell et al., 2013).
In relation to the arguments put forward in various previous studies on the objectives that a hotel chain may set with respect to company growth and the reduction of costs, it seems that, within the strategy of the company, more importance is attached to objectives related to possible association with another chain and to reduced investment of resources with the aim of achieving economies of scale, better control over the reservation systems, and increased company size. In contrast, there is another different tendency to avoid developing forms of closer association with other chains and instead to make greater investments of resources with a view to achieving greater control over the staff training, quality and image of the company, in relation to the objectives that the chain may set for improving its image and differentiating itself from competitors (Contractor and Kundu, 1998a;Ramón, 2000Ramón, , 2002Berbel, 2006;Martorell et al., 2013). Therefore, in this context we propose the following two hypotheses: Hypothesis 4: The perceived strategic importance of the objectives of growth and cost reduction is associated negatively with the mode of entry. Hypothesis 5: The perceived strategic importance of the objective of differentiation is associated positively with the mode of entry.
The theory of real options suggests that, under conditions of high uncertainty, a company will prefer to reduce to the minimum the initial investment while ensuring that it has the option of investing more at a later time, when it will have obtained more information to better reevaluate the uncertainties actually faced (Rivoli and Salorio, 1996;McGrath, 1997;Brouthers et al., 2008). By doing so, the company obtains access (i.e. ownership) to knowledge about the opportunities of investment and about the related uncertainty. In this way the company can restrict the possibilities open to competitors (by closing distribution channels eliminating possible partners, for example), and gains the opportunity (but not the obligation) to adjust the investment position if the uncertainty is reduced or considered manageable. Therefore, we can state that a company will confront uncertainty by choosing the modes of entry that gives it valuable future options (Sánchez, 1993;McGrath, 1997;Folta, 1998;Dixit and Pindyck, 1994;Brouthers et al., 2008). This is embodied in the next hypothesis: Hypothesis 6: The mode of entry is associated negatively with the availability of real strategic options in the organization.
Real options are a clear source of strategic flexibility by capturing the implicit value of the flexibility available to the management. Similarly, these options aid the company management in taking advantage of the opportunities for developing the businesses that are presented by a particular strategic investment. Real options allow a company to abandon or sell the investment project before completing it; to change its use or its technology; to prolong its life; to choose one or other capacity of an investment in plant; to make the investment in R&D more flexible; and to postpone, expand, contract, close and even alter a project during its life (Trigeorgies, 1996;Copeland and Antikarov, 2001;Tamayo, 2006;Li, 2007). Thus, real options are presented as a source of strategic flexibility for an organization, since they allow the decisions taken to be adapted to the real conditions of the business environment (Trigeorgis, 1993;Amram and Kulatilaka, 1999;Copeland and Antikarov, 2001;Tamayo, 2006;Li, 2007;Calle and Tamayo, 2009). We therefore propose the following hypothesis: Hypothesis 7: The availability of real options for the expanding company is associated positively with strategic flexibility, such that the more that real options are open to the company, the more flexible the company will be strategically.
The availability of real options to the company will facilitate the process of organizational change in the face of particular identified contingencies of the business environment. When the directors are aware of these real options, this will give the company valuable strategic flexibility, and should increase the probability that the directors can plan and implement the strategic change necessary for the organization (Tamayo, 2006). Equally, when the company is operating in complex and dynamic settings, strategic flexibility enhances its capacity to respond to the changes in the environment, in the direction required; in other words, it allows the directors to respond when confronted with instability and to develop resources and capabilities to adapt to the changing conditions. If a company is more flexible strategically, its directors will be able to plan beforehand the strategic modifications they consider necessary to adapt to the changes in its environment; this should give the entire company organization greater capacity for adaptation to the requirements of that change that would enable the company to implement successfully the new strategy (Tamayo, 2006).
It is possible to identify two types of result that reflect the international performance of a company: one is the "image and access results", which show its reputation and degree of access to international markets; and the other factor is the "financial results", which represents the financial output obtained by the chain and its degree of satisfaction with the existing global strategy. It should be stated, in addition, that the second factor is a consequence of the first, since the image possessed by a company at the international level and its degree of access to major markets condition the international financial performance of the company (Peterse et al., 2000;Lukas, 2007aLukas, , 2007bSwoboda et al., 2011). To the extent that the existence of real options and strategic flexibility facilitate a good strategic fit, and this in turn leads to better performance, we put forward our last three hypotheses: Hypothesis 8: The availability of real options in the organization is positively associated with access to the markets and the reputation of the organization. Hypothesis 9: The strategic flexibility is associated positively with access to the markets and the reputation of the organization. Hypothesis 10: The results in terms of image and access to the market are associated positively with the financial results of the organization.

Methodology
The population comprises a total of 88 hotel chains, 62 of them Spanish and 26 foreignowned. The geographic scope of the study covers the whole national territory of Spain. The data collection process was carried out between March 2011 and September 2012, by means of a survey of the directors of the leading Spanish hotel chains with international presence and the directors of the foreign chains operating in Spain. The first contact with the chain was made with a telephone call in which the researcher introduced his/herself, and described briefly the purpose of the research, the importance of the company's participation in the study and the confidentiality of the data requested; at the end of this conversation, the prospective respondent was thanked for their collaboration. After that first telephone contact, the next step was to send by mail the survey questionnaire with a letter of presentation and the link to a website from where the whole survey was managed. We employed a questionnaire administered by telephone and personal interviews for the most important chains (by size and international experience).
The questionnaire was completed by 35 of the managers of these hotel chains (22 Spanish and 14 foreign-owned), representing a response rate of 39.8%.
To ensure that the basic characteristics of the target population were maintained in the sample of hotel chains obtained, we verified that there were no significant differences between the population and the sample, based on the structural characteristics of the hotel chains. This was possible because objective data were available on these -specifically on five variables for which we had information at both the population and sampling levels: three referring to the size of the chain and two related to its international experience. Using the SPSS software we applied a test of means to the data.

Trigeorgis
(1996); Copeland and Antikarov (2001;and Tamayo (2006) The company is making an investment in stages, through a series of disbursements; thus it can opt to abandon the project at any time if unfavourable information is received (OREETAP). The company can temporarily close down the investment operations and take them up again when the conditions are more favourable (ORECERR).
The company can abandon an investment project if the market conditions are unfavourable, and can sell the company or a part of it in the second-hand market (OREABAN). Once it has made the initial investment in a project, the company can invest more capital if the conditions are favourable; it is not obliged to make the rest of the investment if conditions are unfavourable (ORECREC). The company commits certain expenditure in a pilot project that may indicate opportunities for growth at some future time based on that project, even though it may not be profitable at present (OREPILO). The company has the option of learning about the market and/or the partners, in strategic alliances, if these provide knowledge useful (OREAPRE). level presents a similar pattern, we concluded that the results could also be generalized, similarly, to other countries and other European hotel chains.

Measurement of the variables
Presented next are the measurement scales used for the various different constructs, summarised in table 1 and explained in the following subsections:

Mode of Entry
Following the studies of Contractor and Kundu (1998a), Ramón (2000), Berbel (2006) and Martorell et al. (2013), we construct the variable M that will be = 1 if the company operates the new foreign establishment under a franchise; 2 = if the company operates it under a management contract; 3 = if under a rental contract; 4 = if under partial/majority ownership (joint venture); and 5= when the company has total ownership and management of the new establishment.

Uncertainty
Although those authors who have analysed the strategy of mode of entry in the hotel industry included some variables of uncertainty in their studies, these are extremely limited, being essentially variables associated with the general environment of the company, related to FLE_EST: Strategic Flexibility (1= in total disagreement; 5= in total agreement) In our company we reformulate the strategies rapidly, when required by the market conditions or the strength of the competition (FEINREFO) Volberda (1997); Verdú (2002) an Tamayo (2006) When the conditions of the environment change we have a variety of strategic measures available to us for coping with that change (FEINMEDI). We use machinery and/or technology for the provision of services which allows us to perform a large number of operations rapidly and without incurring high costs of change (FEINMAQU). The number of modifications introduced each year in the products or services that we offer is high (FEINMODI). In our company we have formal programmes for the orientation of new employees (FEINORIE). In our company we carry out publicity campaigns or promotions with the object of influencing consumer tastes (FEEXPUBL). Our position in the market allows us to monitor the existing competitors and make it difficult for new competitors to enter (FEEXCOMP). In our company we are able to influence certain political decisions intended to modify commercial regulations (FEEXPOLI).
RDO_IMG: Results in terms of Image (1= very much less than expected; 5= very much more than expected)

Costs (RTDOCTES)
Market share (RTDOCUOT) Degree of satisfaction with the entry into external markets (RTDOSATI) the risk of the country (political, economic and financial), to cultural differences, and to the level of political development of the country in which the proposed investment is to be made (Contractor and Kundu, 1998a;Ramón, 2000Ramón, , 2002Berbel, 2006;Martorell et al., 2013). With the object of enhancing the explanatory power of the model, and to be able to test the hypotheses put forward on the relationship of the types of uncertainty (exogenous and endogenous) with the mode of entry chosen by hotel companies, we have adopted a multidimensional conception of uncertainty, in line with the studies of Miller (1992), Werner et al. (1996), Brouthers et al. (2000, 2002, and Sánchez and Pla (2006). Following the recommendation of Li and Rugman (2007), who suggest that, since it is the directors who generate and exercise the real options in international businesses, their perceptions of uncertainty can be an exact measure of this variable, and these perceptions can be obtained by qualitative methods such as interviews or surveys (Li and Rugman, 2007). Therefore, following Brouthers et al. (2000,2002) we have adopted the instrument for measuring perceived environmental uncertainty proposed by Werner et al. (1996). This has been demonstrated to be a reliable instrument for measuring managers' perceptions, as well as being a multiple measure of perceived international uncertainty. This instrument is comprised of 28 items that are grouped together to measure the uncertainty of the policies of the government of a country, the macroeconomics, and the uncertainty of materials, product markets and competition. The two first types of uncertainty are exogenous, while the other three are classified as endogenous uncertainty.
Nevertheless, our proposal is a refinement of the original scale, looking for a better fit to the particular type of companies under study. After an exploratory factor analysis, we found three components of uncertainty of exogenous character, and five of the endogenous one, as shown in Table 1 and, afterwards, in Figure 2. A Likert scale of five points is utilised, and the degree to which each element of uncertainty can be forecasted is evaluated for the particular country into which the hotel company interviewed had most recently entered.

Structural Variables of the Company
Following the studies of Contractor and Kundu (1998a), Campa and Guillén, (1999), Ramón (2000Ramón ( , 2002, Randoy andDibrell, (2002), Berbel (2006), Quer et al. (2007) and Martorell et al. (2013), we use the average sales revenue of the chain in the last three years (2009 to 2011 in our case) on a global basis. We include two additional indicators that reflect the average number of employees in the last three years (Azofra and Martínez, 1999;Rialp et al., 2002;Nakos et al., 2002;Brouthers and Brouthers, 2003;Martorell et al., 2013) and, based on the criterion used to produce the ranking of hotel chains published by Hosteltur and Hostelmarket (professional publications of recognized prestige in the sector), a measure of the total number of rooms that the hotel chain has available, both indicators on a global basis (Martorell et al., 2013). The values obtained, being scalar variables, are transformed into a Likert-type ordinal scale from 1 to 5; for this, 5 equal intervals are calculated in function of the minimum and maximum values of the sample.
The international experience of the company is measured by the number of years since the company opened its first hotel in a foreign country (Contractor and Kundu, 1998a;Ramón, 2000Ramón, , 2002Luo, 2001;Berbel, 2006;Martorell et al., 2013). However, as these authors state, there is a need to monitor the effect of the different kinds of international experience on the mode of entry selected, since international experience can be acquired over the years that the company has been operating in foreign countries or by managing such operations with greater frequency or intensity, over a shorter period of time. Taking this distinction into account, for this variable we include two parameters: first to measure the company's more general international experience, we use the number of years elapsed since the opening of the first hotel abroad; second, in accordance with the preceding reasoning, we also use the current number of hotels the company has abroad (that is, in countries other than the company's country of origin). A third parameter used to measure international experience is the number of foreign countries in which the chain operates, with the object of reflecting how a hotel chain with, for example, five hotels in five different countries ought to have greater international experience (i.e. be more internationalized) greater than another with five hotels in one foreign country. We thus assume that a chain present in a larger number of countries is more internationalized than one present in fewer countries (Caves and Mehra, 1986;Ramón, 2000Ramón, , 2002Martorell et al., 2013). As with the values referring to the size of the chain, those referring to its degree of international experience have also been transformed into a Likert-type scale from 1 to 5, with 5 equal intervals in function of the minimum and maximum values attributed to the parameter.

Strategic and Control Factors
These factors are obtained from the questionnaire completed by the company directors, who were asked to weight the importance of the factors already mentioned, on a semantic differential scale of 5 points, for the world hotel industry, following the studies of Contractor and Kundu (1998a), and for the Spanish hotel industry, following Ramón (2000Ramón ( , 2002, Berbel (2006) and Martorell et al. (2013).

Real Options
Among the various different criteria employed to classify the real options, the perspective based on the flexibility that the option provides to its users is the approach we have adopted (Trigeorgis, 1996;Copeland and Antikarov, 2001;Tamayo, 2006). A semantic differential scale of 5 points is used in a series of items to measure if the company has available to it the real options identified in the specialist literature, in which seven strategic real options can be found.

Strategic Flexibility
To measure this variable, among the different classifications of strategic flexibility that have been produced, we base our choice on the physical orientation, and distinguish between the internal and external aspects of the company organisation. Internal strategic flexibility is related to the capacity of the management to adapt the company to the demands of its business environment, whereas external strategic flexibility refers to the capacity of management to influence that environment, with the object of making the company less vulnerable to changes in the environment. We use a semantic differential scale of 5 points, composed of 8 items (Volberda, 1997;Verdú, 2002;Tamayo, 2006).

International Performance
Focusing on the case of the hotel industry, the study of Berbel (2006) analyses the influence of the mode of entry (as a strategic decision) on the export performance of the international hotel chains; this author develops a scale of measurement for this variable based on a meta-analysis of the literature on export performance. Berbel (2006) evaluates the international performance from an instrument composed of several items. In effect, this author conducts a double survey employing two semantic differential scales to measure the export performance: one for a specific establishment and the other for the hotel chain as a whole (in this case rating the degree to which the actual achievement of objectives compares with the initial expectations). In our study, since our aim is to explain the global result of the international expansion strategy of the hotel chain, we consider that the second measurement is better adapted to the objectives of our study; we also adopt it so as to be able to make comparisons between the results obtained.
However, it should be borne in mind that, within this scale, it is possible to identify the two types of result that a company can obtain with its international strategy, and that reflect its international performance. In this context, by means of an exploratory factorial analysis, we define a first factor as "Result in Image and Access", comprised of the items referring to the reputation of the hotel chain and its degree of access to international markets; and a second factor, "Financial results", comprised of the rest of items that reflect the commercial results obtained by the chain and its degree of satisfaction with its global strategy (Berbel, 2006).

Data Analysis
Due to the size of the sample, 35 valid observations, we have applied the Partial Least Squares technique, since the assumed minimum number of observations for it is 30 (Chin, 2000;Roldán and Cepeda, 2004). We applied the following techniques: -Tests for uni-dimensionality of the scales (factor analysis).
-Analysis of the measurements models: individual reliability of the items (loadings), constructs' reliability (Cronbach's alpha, composite reliability), convergent and discriminant validity (AVE, Average Variance Extracted).   For the statistical treatment of the data collected, the SPSS Version 19 and Visual-PLS 1.04 bi programs have been used. Table 2 synthesizes the main features of the empirical method used for primary data gathering.

Results
After analysing the unidimensionality of the scales utilized, we redefined the scale of the perceived environmental uncertainty and we eliminated some of the indicators utilized in the scales of the real strategic options, strategic flexibility and international performance.
The graphic representation of how the model is configured is shown in Figure 2.
Subsequently, we analysed the measurement models, and verified the individual reliability of each item, the reliability of the constructs, the convergent validity and the discriminant validity, based on the parameters displayed in table 3.
For the interpretation of the values obtained we follow the criterion proposed by Nunnally (1978), who suggests for the reliability a level of 0.7 in the early stages of the research, and of 0.8 in a later stage when the research is more consolidated. It can be observed that all the constructs have a composite reliability greater than or almost equal to 0.8, with values higher than 0.9 in some of them; thus this requirement is met.
With regard to the convergent and discriminant validity through the Average Variance Extracted (AVE) of the constructs, it can also be observed that all of them meet the condition proposed by Fornell and Lacker (1981), who recommend values higher than 0.5, since at this level at least 50% of the variance of the construct is due to its indicators. Similarly, we have verified the discriminant validity of the constructs and confirmed that AVE of each of the constructs is greater that the squared correlations between that construct and the rest that make up the research model (Fornell and Lacker, 1981); this tells us that each construct is different from any of the others; therefore the constructs that have this property are accepted.
The analysis of the strength of the research hypotheses proposed was carried out, first, by verifying that the values of the regression or β coefficients (corresponding to the relationships between the constructs of the structural model) are above 0.2, and eliminating those that did not reach that value. Then a bootstrap analysis was performed. Figure 3 shows the β coefficients and the Student-T test results (in parentheses in the figure) of all the relationships between the constructs of our model.

Discussion
In the light of the results obtained, we can conclude that when the exogenous and endogenous uncertainty perceived by the company increases, selecting a mode of entry that offers the company a high degree of control of the international investments is less beneficial, and there is a greater probability that  the company would prefer a mode of entry with less commitment of resources (e.g. franchising, licensing or joint ventures) as against full ownership (Buckley and Casson, 1998;Leiblein, 2003;Cuypers and Martin, 2007;Li, 2007). We found a negative relationship between greater uncertainty of the "economic policy" type, which encompasses taxation and monetary policy, as well as price regulations by governments (Li and Rugman, 2007;Brouthers et al., 2008;Fisch, 2008;Chen and Chang, 2012). Therefore, when a hotel chain is faced with high uncertainty of this type, it should opt for a mode of entry that implies low levels of investment, thus creating strategic options for increasing its commitment if and when these sources of uncertainty are resolved.
Focusing on endogenous uncertainty, in our model we propose that when the uncertainty is high and endogenous, a multinational hotel company will prefer a mode of entry with high commitment of resources. The results obtained show a negative relationship between the uncertainty associated with the "potential competition" and the mode of entry to the international market chosen, in a direction contrary to that proposed by our hypothesis (Miller, 1992(Miller, , 1993Brouthers et al., 2000Brouthers et al., , 2002. The explanation for this may be found in the fact that, in a market of intense competition, rapid responses by companies to both threats and opportunities are required, and greater flexibility for adapting to the circumstances of the market without incurring significant costs. Therefore, if considerable uncertainty exists about the future competitive pressure, the mode of entry selected should be one involving a lower commitment of resources.
Looking at the effect that the interaction between company size and its international experience has on the mode of entry to the international market chosen, the high volume of investment that the hotel industry requires means that strategic alliances are a more attractive option . For this reason it is important for the company to have the sufficient prestige and resources to undertake the complex process of entering into an alliance. Logically a more experienced and larger company would be more capable of doing this.
The hotel chains face uncertainty by opting for a mode of entry requiring low initial investment but which would give them options to increase this in the future, since the initial commitment of minimum resources under conditions of high uncertainty should be linked with the option of investing more later, once more information has been obtained to re-evaluate the uncertainties identified (Rivoli and Salorio, 1996;Pennings and Lint, 2000;Peterse et al., 2000;Gilroy and Lukas, 2006;Li, 2007). With regard to the relationship between real options and strategic flexibility, we confirm that there is a positive relationship between these factors; hence we conclude that real options are clearly a source of flexibility (Copeland and Antikarov, 2001;Tamayo, 2006;Li, 2007;Calle and Tamayo, 2009).
We have also observed that those chains that possess greater strategic flexibility achieve a better fit with the business environment in which they operate, improving their reputation and their degree of access to the market. This allows us to accept the hypothesis that strategic flexibility is associated positively with this type of international activity (Hitt et al., 1998;Tamayo, 2006;Calle and Tamayo, 2009). To the contrary, in relation to the hypothesis proposed that the availability of real options in the company is positively associated with access to the markets and the reputation of the company, our results show an influence, but in the opposite direction to that expected. We conclude that, among the types of real strategic option available in the international strategy of a hotel chain, the company in effect creates the future option of abandoning the venture simultaneously with its decision to enter a new market (Rivoli and Salorio, 1996;Pennings and Lint, 2000;Peterse et al., 2000;Gilroy and Lukas, 2006;Li, 2007). However, in the event of exercising that option to withdraw, this would damage the reputation of the company and, obviously, diminish its access to that market. Consequently, the hotel chain must take into account the impact on its corporate and market reputation by exercising the option to abandon a venture, particularly when a strategic asset is involved.
Another possible explanation of this finding is in the negative relationship between the mode of entry and the existence of real options that we have proven; this is because this negative relationship indicates that it is the contractual modes of entry that generate more valuable options, and we know that these modes are also those than carry a greater risk in terms of image and prestige, particularly in the case of franchises whereby the company cedes to the franchisee its corporate image and its knowhow. It is therefore no surprise that these modalities have the potential to damage the reputation of the company (Contractor and Kundu, 1998a;Ramón, 2000Ramón, , 2002Berbel, 2006;Martorell et al., 2013). To understand better the variable RDO_IMG (result in image and access) the direct and indirect effects should be analyzed to see which are stronger, since in this case they are contradictory. This, therefore, remains a line of research for the near future.
Lastly, we find that the "financial results" of the international strategy is a consequence of the "image and access result", since the image of the company on an international scale and the degree of access obtained to the markets condition the international financial results. This must be taken into account when the long term strategic objectives of the company are proposed and agreed (Peterse et al., 2000;Lukas, 2007aLukas, , 2007bSwoboda et al., 2011).

Conclusions
In the present research we confirm that the theory of real options can be employed to explain the internationalization strategy of the hotel industry, and we have incorporated additional factors to the existing body of doctrine on the mode of entry to a foreign market. Our study, therefore, contributes to improving understanding of the decision regarding which mode of entry into foreign markets a company in the hotel industry will take. A dynamic theory is also developed to analyse the effects that this decision may have in the future and its impact on the company's international performance.
The theory of real options contributes to the field of international strategy by introducing a new way of thinking, whereby uncertainty is assumed to imply both risks and opportunities; accordingly, a company can benefit from uncertainty by creating real options to maintain the flexibility necessary to adjust its decisions in response to new information.
With respect to the factors to take into account when deciding the best mode of entry to the international market, companies are usually faced with a high degree of perceived uncertainty about economic policy: this is taken to encompass not only the fiscal and monetary policies applicable in the country in question but also the results of economic restructuring. It is therefore recommended that a company should opt for a contractual mode of entry (franchise, management contract or rental contract) with the object of allowing the company the time and opportunity to learn about the new market, thus reducing the uncertainty and enabling the company to subsequently change the way of operating in the new market, if necessary.
Equally, if there is the perception that the degree of competition in the destination market is intense, or if there is much uncertainty about the future degree of competition, with difficulties in predicting the entry of new competitors, the company should consider entering the market with a mode of entry involving the commitment of minimum resources.
Given the large investment required by the hotel industry, and since there is a positive inter-relationship between size and international experience that facilitates the formation of strategic alliances by means of contractual agreements for entering new country markets, those larger hotel companies with more experience should clearly exploit these competitive advantages to form such alliances. This approach would give them valuable strategic options for learning and possible changing course at less cost, which should translate into an improved international performance for the company. Equally, those chains that do not possess these advantages need to make efforts to acquire them before attempting to expand into other countries. To this end, greater international experience could be gained with a strategy of opening of new establishments in carefully-selected foreign countries, thus increasing the number and variety of countries in which the chain has presence and reducing its geographic concentration that would limit the advantages of multinational operations.
Finally, we also confirm that entering a new market by means of contractual formulas in the face of large uncertainties creates the option of abandoning it if the conditions are worse than expected, with the object of minimizing losses for the chain. However, it is essential for the company to realise that, by exercising this option it may harm its reputation, and reduce its prospects of gaining access to that market in the future and obtaining the advantages of the multinational operation.
Therefore, before deciding to pull out of the unsuccessful venture, the company will need to weigh up the potential costs and benefits: the reduction of future losses should be set against the impact of the decision on its existing multinational network and the reputation of the chain, particularly when this reputation is considered a strategic asset.
To finish this part, returning to the research questions put forward initially, we can state that international uncertainty influences the mode of entry chosen by the hotel chains, in such a way that the greater uncertainty perceived in respect of "economic policy" and "potential competition" has a negative influence on this choice. Equally, the structural variables of the company, that is, the interaction between its size and its international experience also have a negative impact on the chosen mode of internationalization.
Regarding the question of how the strategic factors and those of control of the company influence the volume of investment in international markets, the results obtained have required us to eliminate from the model the postulated relationships between the objectives of growth and cost reduction and the mode of entry chosen.
We have been able to confirm that the choice of mode of entry to an international market influences the real strategic options open to the company. We observe that the hotel chains confront uncertainty by choosing modes of entry of low investment that leave more future options open to them, in such a way that the reduced initial commitment of resources under conditions of high uncertainty assures the option of investing more at a later time (option of change), as well as creating options for learning, for growth or for abandoning the venture. Equally, the existence of real options for the company contributes to its strategic flexibility.
We also pose questions about the possible effects of real options and strategic flexibility on the international performance of the hotel chain. In this respect we have confirmed that greater strategic flexibility allows the hotel chain to adjust better to its business environment, improving both its reputation and its level of access to the market. To the contrary, the option of abandoning the venture can have a negative influence on the image and the multinational projection of the chain, which is a factor to consider before taking this option.
Lastly, we confirm that the "financial results" of the hotel chain is the consequence of the result in terms of its "image and access".

Limitations
The main limitation of our study is the size of the sample (35 responses); in addition, the lack of collaboration by many of the managers of the hotel chains to whom we have addressed our survey, has reduced the size of the population sampled and has lengthened the time needed to obtain the responses; these two factors have conditioned the statistical techniques utilized for the analysis of the research model. Equally, the extended period of time taken to collect the data may have introduced some distortion in the results. On this point we believe that any such distortion would not be very significant, since the evolution presented by the hotel chains in recent years, with respect to the modes of entry employed, does not show large variations over time.
It is clear that we have not addressed the full complexity of the variable «mode of entry». What we have considered, in effect, is a polytomous ordinal measurement that, when it increases it shows that the multinational possesses a greater degree of shareholder ownership and total control over the hotel in the foreign country, in line with the earlier studies applied to the Spanish hotel sector (Ramón, 2000;Berbel, 2006 andMartorellet al., 2013). In this measurement, we have not differentiated certain nuances related to some particular situations or formulas, which could enrich the analysis.
The real options have been identified, but they have not been valued. We have not valued the real options open to the chain; we have only considered whether or not the managers have been capable of identifying them, under the assumption that it is their identification that allows them to be executed and exploited. This is in line with the arguments of Driouchi and Bennett (2011), who show the importance of identifying and adequately exploiting the real options, which would provide greater returns for the multinational chain in the long term, and better returns than its competitors who are not capable of identifying these options.

Future lines of research
Regarding the avenues for possible future research, we commence with the possible applications of the theory of real options to international strategy; its application could be generalized by studying other sectors of industry. One of the first lines of work would be to improve knowledge of the ways in which uncertainty, both exogenous and endogenous, influences the strategic decision-making of a multinational company, thus extending the hypotheses of Li and Rugman (2007), tested in our model, to other industries.
Given that the hotel sector presents peculiarities in respect of the degree of control that the chains can exercise with some of the contractual modes of entry, a second line of research could be to test the direction of the hypotheses of Li and Rugman (2007) for the case of the hotel industry, distinguishing between entry modes that involve control from those that do not, without linking control and ownership, in line with the study carried out by Graf (2009).
Another line of research would be centred on the analysis of the conditions in which the multinational companies, integrated into networks, exercise the real options, the modes and the timing of market entry, identifying the variables with most influence in the strategic fit, in line with the studies of Fisch (2011) and Swoboda et al. (2011).
Equally, more in-depth studies could be made, in line with our model or with that developed by Brouthers et al. (2008), of the ways in which the generation and exercise of real options impact on the performance of a multinational company.
Lastly, in line with the model that we have developed, the number of observations could be increased, to make possible a multi-group analysis comparing Spanish chains with foreign-owned chains. It would also be worth replicating our study after a certain period of time has elapsed, with the object of conducting a dynamic or longitudinal analysis.