Increasing Sustainability through Wine Tourism in Mass Tourism Destinations. The Case of the Balearic Islands

This study explores the potential contribution to economic and cultural sustainability of fostering wine tourism activity in a Mediterranean mature mass tourism destination like the Balearic Islands. A causal model was designed linking winery visitors’ satisfaction with both the local wines and the visit to the winery and future purchasing intentions as well as the interest in the destination cultural offer, global satisfaction with the trip, and image of the destination. Field research was carried out within the wineries where a questionnaire was administered personally to a sample of 200 winery visitors. The managers of the wineries were contacted previously to ensure their cooperation in the research. The statistical method used for the analysis of the exploratory causal model was the Partial Least Squares (PLS) regression. Results indicate that there is a great growth potential for wine tourism activity in the islands with important positive externalities on gastronomic tourism and the overall image of the Islands. Results show that wine tourism might boost the sales and exports of wine in the medium and long term. Finally, it can be concluded that there are important synergies between rural tourism, mass tourism, and wine tourism in the islands which are still not sufficiently optimized, and that greater support to wine tourism development could contribute not only to a more sustainable rural development but also to a more sustainable tourist activity overall within a more diversified economy.


Introduction
Attractive and preserved landscapes are a key resource for the attractiveness of tourism destinations. In this sense, rural activities (agriculture, livestock) can play an essential catalytic role by contributing to preserve and shape the countryside, allowing the visitor to enjoy more beautiful sceneries. Therefore, the more sustainable the agricultural and livestock activities are, the more they can contribute to the sustainability of the tourist activity, by making the tourist destination more attractive and, consequently, more competitive.
Sustainability becomes a priority objective especially in the case of small island territories because of the fragile nature of their ecosystems being absolutely necessary to reach a correct balance between economic development, sustainability, and environmental protection. This meaning that the economic development of rural areas should prevent unwanted social and environmental negative impacts. Moreover, if tourism activity takes place in such small territories, it becomes essential to work to achieve sustainable protection of the different ecosystems (agriculture, food, forestry, and tourism) and the natural and cultural heritage associated with the rural activities. In this regard, innovation and entrepreneurship have become fundamental instruments in rural areas to face current economic, environmental, and social challenges in a sustainable way.
Winemaking is a long historical activity that has traditionally contributed to the sustainability of rural areas in different parts of Europe and worldwide. When referring to (c) What is the influence of the previous relationships on the satisfaction with the cultural offer of the destination and the global satisfaction of the visit to the destination; (d) The influence of the visit to the wineries on visitor's future intentions to purchase local wines either during the trip or when returning home; (e) And, lastly, the impact of the previous relationships on the image of the tourist destination.
The ultimate goal of this research is to explore the potential contribution of fostering the development of wine tourism to the global sustainability of both, the rural areas and the tourism destination as well and the positive synergies that could emerge.

Winemaking and Wine Tourism in a Mass Tourism Destination
The Balearic Islands is an archipelago in the Western Mediterranean, with an area of 4992 km 2 and over 1,100,000 inhabitants [21]. The islands are a worldwide-known tourism destination relying mainly on 'sun and beach' tourism and receiving more than 13 million tourists annually, mostly concentrated on the summer season from April to October [22]. Tourism activity is absolutely prevailing and accounting for more than 65% of the island's GDP. Tourism began in the Balearic Islands at the beginning of the 20th century but the tourism boom occurred in the 1960s and 1970s with extraordinary annual increases in visitor arrivals, leaving the other economic sectors in a secondary position, including the agricultural sector. One of the objectives of the tourism policy of the Islands during the last decades has been to reduce the seasonality of the tourist arrivals due to its rather unsustainable impact on the physical environment and its adverse economic and social effects [23]. This implying diversifying the touristic offer and seeking to attract alternative tourism segments with different seasonal behavior and different profiles than the current predominant tourism, like cultural tourism. Gastronomic tourism is a targeted key sub-segment within cultural tourism being wine tourism a very specific part of gastronomic tourism.
Although wine production in the Balearic Islands dates back to the times of the Roman Empire, most of the wineries are of very recent creation and wine tourism activities go back no more than a decade being this development exclusively due to the entrepreneurship of a new generation of winemakers and without any support from the Public Administration. Figure 1 illustrates when the wineries participating in the study were founded.

Research Background
In the past 30 years, the tourism markets have experienced important changes with tourism demand experiencing a progressive process of greater diversification. Thus, we find a market where travelers have increasingly different motivations and are seeking different benefits in their holiday-trips. Travelers that take shorter and more frequent trips which, besides visiting the traditional sun and beach destinations, also involve finding new places and experiences. One of these trends is the interest shown in knowing everything related to the food and wine of the region visited [24,25].
Wine tourism is often associated with cultural tourism, rural tourism, and gastronomic tourism [3], as they are complementary offers. Coros, Pop & Popa [26] also categorize wine tourism as a subtype of gastronomic tourism stating that wine, food, and cuisine represent a tourist vector and a cultural component. The link wine tourists establish between wine and local cuisine is also highlighted by [27] when analyzing motivations and perceptions among wine tourists in the Sherry region of Spain.
Some previous research on wine tourism, from the demand perspective, has concentrated on the impacts of the winery visits on future behavioral intentions, focusing on two main aspects: revisit intentions to the winery and future wine purchase intentions. Also, the destinations' image, their most prominent dimensions for consumer's choice and the link between the visitor's involvement in wine and the image of the wine region visited have been the object of research analysis by some academics [28][29][30].

Satisfaction, Revisit, and Future Purchase Intentions
Several studies have analyzed the influence of winery visits on the future behavioral intentions of visitors, particularly, purchase intention. Satisfaction and positive emotions derived from the visit have been constructs frequently analyzed. In this regard, [31] concluded that tasting sessions for tourists at Spanish wineries were a direct way to promote wines and encourage purchases. More precisely, these authors conclude that the positive emotions generated by the product's novelty (that is, the novelty of the wine itself) highly explained purchase intention, whilst the emotions produced by the winery visit barely affected it. Therefore, the wine itself has greater importance than the wineryvisit experience as an antecedent of wine-purchasing intentions. Bruwer, Coode, Saliba & Herbst [32] slightly shifted the focus to brand attitudes and loyalty reaching similar conclusions in that the quality of the wine was the most influential factor to increase the likelihood of future purchase intentions. However, these authors placed more emphasis on the winery visit experience stating that consumers having an enjoyable and memorable experience are more likely to buy the wine again and/or promote the wine brand to others.
Finally, [33] concluded that the intrinsic components of wine tourism products that satisfy today's experienced travelers and, hence, are critical to developing wine regions, should focus on the design of activities with a significant dose of local identity and wine and gastronomic pairing. Therefore, they highlight the role of local culture in the satisfaction of today's experienced travelers who are seeking new and memorable experiences developed within authentic contexts.
Leri &Theodoridis [34] also focused their research on the winery visit experience analyzing its impact on the visitor's emotions and the effect of satisfaction on visitor's intention to revisit and recommend a winery, concluding that the winery environment is an important factor for visitor's satisfaction which, in turn, is a significant predictor of visitor's intention to revisit the winery and give positive word-of-mouth. Back, Bufquin & Park [35] also studied the factors influencing revisit intentions concluding that the reputation and perceived quality of the winery and its wines significantly affect the number of visits and revisit intentions.

Cultural Image and Visitor's Involvement with Wines
Academics have also concentrated research efforts studying the image of wine tourism destinations and their most important dimensions acting as pull factors for destination choice. In this regard, [28] and [29], in the context of Canadian and US wine regions respectively, reached the conclusion that the destination's natural beauty and geographical setting of its landscape were the most important image dimension for destination choice.
Alternatively, other authors [30,36] have focused on the relationships among the visitor's wine product involvement and the wine region image as a tourism destination and how this could affect future behavioral intentions, analyzing both actual and potential visitors, In this regard, [36] reached the conclusion that some positive links do exist between wine product involvement and the perceived destination's cognitive and affective image among potential visitor's while [30] concluded that consumers who were highly involved with wine appeared to be more influenced by affective destination image and self-congruity than those who indicated low wine involvement. Both studies conclude that, indirectly, wine tourism was favoring visits to some regions strengthening their image as cultural tourism destinations.
However, no studies have been identified exploring the potential positive impact of wine tourism activities on the cultural dimension of the image of a tourism destination whose image is overwhelmingly dominated by aspects like 'good climate, sun and beach and fun' as is the case of the Balearic Islands and where Destination Marketing Organizations (DMOs) are trying to systematically complement this image by strengthening the cultural dimensions associated to it, as a way to break seasonality and, therefore, increase the sustainability of the tourism model.
Only [15] analyzed the potential of wine tourism in the context of a mass tourism destination that shares many similarities with the Balearic Islands, i.e., the Canary Islands whose tourism industry is also centered around competitively priced holidays focusing on the sun and beach mass tourism experience. Although their study examined the potential of wine tourism from the winery operators' perspectives, it concludes that the restructuring of the islands' wine industry could offer opportunities for developing new tourism alternatives based on gourmet products and traditional landscapes. The authors also pointed out the main challenges to overcome in the development of a successful sustainable local wine tourism industry which included the need for expansion of the destination image to reflect the region's wine-making history and scenic qualities, a shift towards independent high-yield travelers, and reintroducing local produce in the mass tourism product.

Research Design
Based on the previous research background, an exploratory study has been proposed. Specifically, an exploratory Structural Equations Model (SEM) was created ( Figure 2). The exploratory model proposes a set of direct causal relationships between various elements of the tourist experience. The causal relationships were proposed on the basis of the literature, the authors' experience and comments from winery managers obtained in pre-field interviews. Moreover, the elements of the model are important constructs in the analysis of tourist behaviour. These constructs were measured with reflective item scales, a common technique in consumer behaviour analysis. The objective is to determine which causal relationships have some kind of support. The following set of causal relationships were framed to measure the impact of winery visits on the visitor's satisfaction and future behavioral intentions (regarding wine purchasing and holiday behavior during the rest of their stay in the destination) as well as the impact on the image, in the context of a consolidated mass tourism destination, as is the case of the Balearic Islands. Causal Relationship 1: Visitor's interest in wines has a direct effect on satisfaction with local wines. Causal Relationship 2: Visitor's interest in wines has a direct effect on satisfaction with the winery visit. Causal Relationship 3: Satisfaction with local wines has a direct effect on satisfaction with the winery visit. Causal Relationship 4: Satisfaction with local wines has a direct effect on interest in local gastronomy. Causal Relationship 5: Satisfaction with local wines has a direct effect on the intention to buy wine on the island. Causal Relationship 6: Satisfaction with the winery visit has a direct effect on interest in local gastronomy. Causal Relationship 7: Satisfaction with the winery visit has a direct effect on the intention to buy wine on the island. Causal Relationship 8: Satisfaction with the winery visit has a direct effect on the intention to buy wine after the trip. Causal Relationship 9: Interest in local gastronomy has a direct effect on satisfaction with the local gastronomy. Causal Relationship 10: Satisfaction with the local gastronomy has a direct effect on the intention to buy wine on the island. Causal Relationship 11: Satisfaction with the local gastronomy has a direct effect on satisfaction with the destination's cultural tourism offer. Causal Relationship 12: Satisfaction with the destination's cultural tourism offer has a direct effect on the intention to buy wine on the island. Causal Relationship 13: Satisfaction with the destination's cultural tourism offer has a direct effect on satisfaction with the trip to the island. Causal Relationship 14: Satisfaction with the trip to the island has a direct effect on the intention to buy wine after the trip. Causal Relationship 15: The intention to buy wine on the island has a direct effect on the intention to buy wine after the trip. Causal Relationship 16: Satisfaction with the trip to the island has a direct effect on the destination image of the island.
Causal Relationship 17: Satisfaction with the winery visit has a direct effect on the destination image of the island.

Sampling and Data Collection
For the measurement and testing of the proposed causal relationships, a questionnaire was used to analyse various constructs measured through items created from an initial pre-test. In this pre-test there were item scales for each construct, elicited and adapted from the literature, the authors' experience and interviews with winery managers. A 5-point Likert scale was used for the answers to the various items, following the recommendations of [37], with 1 "Strongly disagree", 2 "Disagree", 3 "Neither agree nor disagree", 4 "Agree", 5 "Strongly agree". A pilot test was carried out to identify possible weaknesses and lack of understanding of the questionnaire. The questionnaire was completed with questions regarding the reasons for the visit to the island (Table 1), the characteristics of the trip and the visit to the winery (Table 2), and demographic profile ( Table 3). The questionnaire was administered personally to a sample of 200 winery visitors. All interviewees answered all questions, with some questions being multiple choice. The managers of the wineries were contacted previously to ensure their cooperation in the research.    Winery visitors in the Balearics show, in general terms, the same profile of nationalities as traditional tourism in the Islands although the percentage of Swiss and Scandinavians is relatively higher. However, the socioeconomic profile of winery visitors is substantially different: they are middle-aged, with high levels of education, and with medium-tohigh incomes. Regarding the characteristics of the visitors to the wineries, most are couples traveling on their own and whose main source of information is the Internet. They travel for vacations during the summer months looking for the attractiveness of the climate, but they 'go a step further-on' and are seeking additional and alternative leisure experiences. Currently, wine tourism is a complementary activity but not the main motivation for the trip.
The statistical method used for the analysis of the Exploratory Causal Model was the Partial Least Squares (PLS) regression. In particular, the SmartPLS software [38] was used for the analysis. The objective of PLS modelling is the prediction of the dependent variables, both latent and manifest. This technique, compared to methods based on covariance (CBM), has the advantage of simultaneously estimating all coefficients and individual variables in the context of a specified model and, as a result, avoids the estimation of inconsistent and partial parameters [39].
PLS is best suited for predictive applications [39] and theory developments (exploratory analysis, as in this case), while CBM is better for applications that seek confirmation of theory (confirmatory analysis). If the work to be carried out was a replication of small models analysed in previous studies in a new sector or destination, methods based on covariance would be the best option [40,41], but as a significant extension is made which implies that we are facing a theoretical development, methods based on variance are better. This leads us to conclude that the use of PLS-based methods is the right option for the analysis proposed since we are dealing with a complex causal-predictive analysis [39][40][41][42][43][44].
Furthermore, PLS follows a complex model segmentation treatment and therefore can work with small sample sizes. Since it consists of a process of estimating subsets in single and multiple regressions, the required sample will be the one that serves as the basis for the more complex multiple regression present in the causal model [45]. Using this method of calculation, the minimum sample size required to test causal relationships with perfectly independent variables and four predictors, with a power of 0.80 and an Alpha of 0.05 (probability of a false positive), is 84 cases to detect factorial loads or Betas equal to or greater than 0.30, according to the approximation to the tables of [46] developed by [47]. Samples greater than 100 are recommended but in this case a sample of 200 was targeted to detect factorial loads or causal effects of a size smaller than 0.30. Larger sample sizes allow greater reliability and precision, increasing the consistency of the results.

Results
First, some results derived from the analysis of the questions regarding the demographic profile of the visitors and the visit itself can be indicated. Although visits to the wineries are seasonal they are distributed more homogeneously and throughout a longer period than the traditional and dominant "sun and beach" tourism ( Figure 3). However, it should be noted that, according to our estimates based on the data collected from the winemakers who collaborated in the study, the volume of visits to the wineries is equivalent to 1% of the total number of tourist visitors to the archipelago. Therefore, the ability of wine tourism to reduce the seasonality of tourism flows to the Islands is limited as long as the wine tourism component of the island's image is not more strongly projected.
Another interesting result is the correlation found between the country of residence of winery visitors and the country of destination of Balearic wine exports. Only 23% of Balearic wine production is exported, the rest being consumed within the islands. As it can be observed in Figure 4, the main destination market for Balearic's wine exports is Germany, where the majority of visitors to the wineries are from. The United Kingdom, in a negative sense, and Switzerland, in a positive sense, are the only significant exceptions to this correlation.

Measurement Model Analysis
In this study, scales created expressly have been used and, therefore, they had not been tested in previous studies. Although the previous tests allowed to obtain guarantees on the designed scales, there was not total security on these scales before their final application. After fieldwork, various checks of the measurement model were carried out: Reliability, Convergent Validity, and Discriminant Validity.
The criterion usually adopted to contrast the individual reliability is that the items have a loading or simple correlation greater than 0.707 [49]. Although most of the items finally conserved in the measurement model fulfil this criterion to a large extent, some items with values very close to this criterion have been conserved because it is a study with a large exploratory component.
The construct reliability measures the indicators' consistency that makes up each of the constructs, that is, that the indicators or manifest variables are measuring the same concept or latent variable. The two most used reliability indicators are Cronbach's Alpha [50] or individual reliability, and composite reliability [51,52]. Both statistics range from 0 (null homogeneity) to 1 (total homogeneity). The Cronbach's Alpha assumes that each indicator contributes in the same way and the composite reliability uses the item loadings determined in the causal model. The most common is to consider as a reliability criterion that the values of both indicators are higher than 0.7, although it is much better if it is higher than 0.8 [53]. Individual reliability is low in three of the constructs used, but composite reliability is good or very good in all cases (Table 4).
Convergent validity [51] implies that a list of indicators represents an only underlying construct, and this can be checked through its unidimensionality [54]. It is usually evaluated using the Average Extracted Variance (AVE) and the minimum recommended values are 0.5 [52]. In this study, all the constructs used meet the AVE criteria (Table 4).
Discriminant validity [51] implies that each construct must be different from the rest of the constructs. The two methods for its valuation are: loadings must be greater than crossloadings; AVE square root must be greater than the correlations with other constructs [55]. When analysing the table of cross-loadings (Table A1 in the Appendix A) and comparing the correlations with the AVE square root (Table A2 in the Appendix A) it is observed that the discriminant validity is fulfilled in the constructs used.
The analysis of the measurement model has forced the elimination of several items due to reliability and validity problems. Specifically, the variables have been eliminated: "I have learned about wine tasting" from "Interest in Wines"; "Balearic wines have a good quality-price" from "Satisfaction with Local Wines "; "The Island's wineries have an adequate range of activities" of "Satisfaction with the Winery Visit"; "The Island has a great abundance of historical monuments" of "Satisfaction with Cultural Offer". As a result, the items indicated in Table 5 have been preserved.

Structural Model Analysis
In the structural model analysis, the relationships that appear reflected in the causal diagram are contrasted. In essence, the standardized beta regression coefficients, the Student's T values, and the R 2 are obtained. The regression coefficients allow us to analyse the fulfilment of the proposed causal relationships and it is desirable that they are above 0.3. The R 2 measures the amount of variance that is explained by the model proposed in the causality diagram and it is advisable to focus on the study of those greater than 0.1 [56].
PLS does not assume the normal distribution of the data, which means that the parametric significance tests are not reliable. Instead, PLS relies on a nonparametric bootstrap procedure [57,58] to test the significance of the estimated regression coefficients. Bootstrapping is a resampling process in which N samples (in this case 5000) are randomly generated from the original sample by substitution with replacement. The parameter estimates generated from the subsamples are used to obtain the standard errors, and the T values are calculated to evaluate the significance of each estimate [59]. The results of the analysis of the significance of the different causal relationships of the proposed causal diagram, both through the use of Student's T values and through non-parametric techniques [54], are detailed in Table 6. As can be seen Table 6, there is a positive link between visitors' interest in wines and the satisfaction with local wines (CR 1 ). In turn, satisfaction with local wines has a strong positive influence on the satisfaction with the winery visit (CR 3 ) and future purchasing intentions of local wines either during the stay on the island (CR 5 ) or after returning home (CR 15 ).
Satisfaction with local wines has a direct effect on interest in local gastronomy (CR 4 ) which is positively linked with satisfaction with the local gastronomy (CR 9 ). In turn, satisfaction with local gastronomy has a direct effect on satisfaction with the destination's cultural tourism offer (CR 11 ) which strongly positively influences overall satisfaction with the trip to the island (CR 13 ). Finally, satisfaction with the trip to the island has a positive effect on the destination image of the island (CR 16 ), a logical and expected effect.
Results also confirm CR 8 (satisfaction with the winery visit has a direct effect on the intention to buy wine after the trip) and CR 14 (satisfaction with the trip to the island has a direct effect on the intention to buy wine after the trip), although these relationships are the less intense within the exploratory model. These results indicating that the quality of wines plays a more salient role in shaping future purchasing intentions than the winery visit experience. The low intensity of CR 14 indicates that these are constructs with little connection to each other, that is, satisfaction with the trip to the island may exercise a positive influence on revisit intentions but not on future purchasing intentions, something that is basically influenced by the visitor's perception of the quality of wines.
Six causal relationships are not significant, so we cannot affirm that such relationships exist. Therefore, we cannot affirm that: the interest in wine has a direct effect on the satisfaction with the winery visit (RC 2 ); satisfaction with the winery visit has a direct effect on the interest for local gastronomy (RC 6 ), on the intention to buy wine on the island (RC 7 ), nor on the destination image of the island (RC 17 ); satisfaction with gastronomy has a direct effect on the intention to buy wine on the island (RC 10 ); Satisfaction with the cultural tourism offer has a direct effect on the intention to buy wine on the island (RC 12 ).
The results of the analysis of the exploratory model and the 17 causal relationships proposed can be summarised in four main ideas: 1.
There is a strong relationship between satisfaction with local wines and interest for local gastronomy (CR 4 ).

2.
Tourists with more interest in wines and gastronomy are more satisfied with local wines and gastronomy (CR 1 , CR 9 ). When there is no interest, satisfaction will be low regardless of the quality and originality of the good, service or experience.

3.
Satisfaction with local wines increases the intention to buy local wine, both during and after the trip (CR 5 , CR 15 ). If the tourist tastes the local wine and likes it, he/she is very likely to buy a bottle as a souvenir of the trip and will try to buy wine from the region once he/she has returned home.

4.
Interest and satisfaction with local gastronomy (including wine) is a key element for visitors' satisfaction with the destination's cultural offer (CR 9 , CR 11 ) and indirectly increases overall satisfaction with the trip and enhances the destination image (CR 13 , CR 16 ). Sedmak and Milhalic [60] found that local gastronomy and the natural environment are central to the perceived authenticity of the tourism experience and thus to overall trip satisfaction. In essence, gastronomy is a fundamental pillar for tourists interested in learning about the local culture.

Discussion
The results obtained in this study are, thus, in line with those obtained in the few previous studies identified that have studied similar relationships but with some differential nuances. The results confirm that: (a) Future purchasing intentions (either while on holidays or after returning home) are predominantly influenced by the quality of the wines tasted during the winery visit and, to a lesser extent, by the winery experience itself. Therefore, the results are more or less aligned with the findings obtained by [31] in the context of a well-known Spanish region and wineries who concluded that the positive emotions generated by the novelty of the wine itself significantly explained purchase intention, whilst the emotions produced by the winery visit barely affected it. Although to a lesser extent, they are also in line with the results obtained by [32] who reached a similar conclusion but placed more emphasis on the winery visit experience stating that consumers having an enjoyable and memorable experience are more likely to buy the wine again and/or promote the wine brand to others. Our results show that is the satisfaction with the wines tested is the most prominent factor explaining future purchasing intentions of wine while also having a strong positive effect on satisfaction with the visit to the winery, but the latter has a lesser, albeit positive, effect on the future intention to buy wine. Therefore, our results would reveal an intermediate position in comparison with the results obtained by these authors. Another aspect that should be noted concerning this issue, and considering the results obtained by the two studies previously mentioned, is that consumer or visitor's behaviour seems to be very similar, no matter if the winery visit takes place in a region widely known for its wines or in a mass tourism destination that consumers do not associate with wine. (b) The visit to the wineries has a significant and positive effect on strengthening the cultural dimension in the image of a consolidated tourist destination in whose image the components associated with 'sun and beach holidays' predominate. Therefore, our results are, to a certain extent, aligned with those obtained by [30] and [36] who concluded that, indirectly, wine tourism was favouring visits to some regions strengthening their image as cultural tourism destinations. Our study also indicates that the satisfaction with local wines reinforces the winery visitors' interest in and satisfaction with the local gastronomic offer which, in turn, has a positive effect on their satisfaction with the cultural offer of the destination visited. Crespi-Vallbona & Mascarilla-Miró [33] were also highlighting the wine and gastronomic pairing and stressing the role of local culture in the satisfaction of today's experienced travellers. (c) Finally, and derived from the links discussed above, wine tourism has a very positive influence on the overall satisfaction with the trip which, in turn, has a strong and positive impact on the image of the tourism destination reinforcing their cultural dimension. The aforementioned effects signal to an important potential contribution of wine tourism to the economic and cultural sustainability of the islands.
In terms of theoretical implications, it can be stated that the results of this research (carried out in the very specific context of a mass tourism destination widely perceived by the main European outbound tourism markets as a summer destination with an offer mostly focused on the enjoyment of the good climate, the sun, the sea, the coastline, and the beach) do not differ much from the results obtained in areas specifically devoted to wine tourism. Therefore, the results while reinforcing the findings obtained in previous studies which, directly or indirectly, have analysed similar topics as this research, indicate that these patterns of behaviour among winery visitors could be generalized to any geographical and/or tourist context.
Gnoth [61] presented a theoretical and practical model illustrating how the development of a country as a tourism destination brand created leverage for its products and services in export markets. In our study, we took the opposite approach by showing that encouraging the testing of a particular product during the holiday stay, in this case, wine, has a positive effect on the tourism destination brand, improving its image while favouring exports. This strengthening of the cultural dimension of the destination's image would have a favourable effect on the objective of reducing the high seasonality of demand flows, an objective which, as mentioned above, is among the most pursued by the Balearic Islands' tourism policy.

Conclusions and Practical Implications
The results of this study point to high levels of satisfaction with local wines among those tourists visiting wineries while spending their holidays in the Balearic Islands and clearly interested in wines in their daily lives. The results also indicate a high level of satisfaction with the wineries visit experience. Therefore, there is an interesting potential for further development and growth of wine tourism in the islands. Scherrer et al. [15] reached a similar conclusion in the context of another summer mass tourism destination, the Canary Islands.
Wine tourism in the Balearic Islands can make an important contribution to rural development, benefiting areas outside of the coastal tourist zones with a more socially and environmentally sustainable footprint in comparison to mass tourism. Therefore, if wine tourism was more strongly promoted, this could contribute to more sustainable and balanced development in the islands, both in strictly economic, social, and environmental terms.
The contribution to increased sustainability would stem from the positive impact on the overall image of the islands reinforcing their cultural dimension, as proved by the results of this study. This fact could foster the demand for cultural tourism in the islands with positive effects on lowering seasonality as cultural tourism shows a less seasonal demand pattern than summer-sea holidays. Furthermore, the incipient demand for wine tourism in the Balearic Islands is already showing a less seasonal pattern than overall tourism demand in the islands.
Within the same context of the Balearic Islands, Serra et al. [62] were pointing out that rural tourism was contributing to improving the overall image of the islands as images of the preserved rural-landscapes and the heritage associated with rural activities were intensively used by the institutional promotion carried out by DMOs (Destination Marketing Organizations), thus, concluding that clear synergies could be identified between both types of tourism in the islands, i.e., rural tourism and mass tourism, considering that rural tourism development has contributed to preserving this countryside scenarios and rural heritage. This synergetic effect could be reinforced by stimulating wine tourism in the islands.
Additionally, our results indicate that fostering wine tourism would have a positive impact in attracting tourism segments more interested in culture and gastronomy. The pairing between wine tourism and local gastronomy has the potential to increase demand for specific gastronomic tourism offers which could promote the demand for local food, produced in the islands (this is increasingly demanded by more sophisticated and more environmentally aware consumers). Not to mention the direct positive impact in fostering the demand for export wine shown by this research that could be derived by encouraging wine tourism.
Altogether, it could contribute to the revitalization of agricultural activities, which have been one of the sectors most negatively affected by the intense tourist development experienced by the islands over the last 50 years. This adverse 'collateral effect' being due to the fact that large hotel establishments have prioritized price when providing food supplies, placing considerations relating to the sustainability of the rural economy in second place, even though the preservation of the rural countryside was important for the attractiveness of the tourism destination. These weak relationships and symbiosis between the local farming and hospitality sectors were also highlighted by [16] in the context of the Canary Islands when indicating that the local hospitality industry was favouring cheaper products from outside the archipelago over those grown/produced locally. An additional aspect showing parallelism between some developments in both archipelagos. Thus, successful developmental synergies of tertiary and primary sectors are still in question in places that have experienced rapid and massive tourism development based on summer coastal tourism. Wine tourism development could boost synergies between these two sectors partially compensating for the damage the development of tourism caused to the primary sector in the Balearics.
Overall, it can be stated that there is a great growth potential for the wine tourism activity in the islands with important positive externalities on gastronomic tourism and the overall image of the Islands. However, wine tourism proves to be important not just because it increases the wineries' short-term income, due to the tourists' expense during the visit to the wineries but also because it boosts the sales and exports of wine in the medium and long term. The nascent wine tourism activity in the islands has been, up to now, the result of the entrepreneurship of new generations of wine producers without any kind of institutional support. Insufficient economic incentives were also pointed out by Gázquez-Abad et al. [11] as having a significant influence on sustainability strategies by Spain wineries Therefore, to benefit from the opportunities arising from the development of wine tourism in the islands, actions should be implemented by the Public Administrations to liberalize, simplify and facilitate the production of wine and promote wine tourism. The cooperation between private businesses and the public government would also facilitate innovation and the development of creative and successful wine tourism offerings.
Finally, it can be concluded that there are important synergies between rural tourism, mass tourism, and wine tourism in the islands which are still not sufficiently optimized and that greater promotion of wine tourism could boost them and contribute not only to a more sustainable rural development but also to a more sustainable tourist activity overall within a more diversified economy.
Research limitations. As results have been obtained in a specific mass tourist destination, it is suggested to replicate the study in other destinations with similar characteristics to gather additional information allowing to generalize the findings. Although the proposed exploratory model incorporates multiple constructs and causal relationships, there might be variables and causal links not considered having a significant effect on the behavior of wine tourists.

Conflicts of Interest:
The authors declare no conflict of interest.