**Developing a Competitive and Sustainable Destination of the Future: Clusters and Predictors of Successful National-Level Destination Governance across Destination Life-Cycle**

#### **Ivan Paunovi´c 1,\*, Marc Dressler 2, Tatjana Mamula Nikoli´c <sup>3</sup> and Sanja Popovi´c Panti´c <sup>4</sup>**


Received: 15 April 2020; Accepted: 13 May 2020; Published: 15 May 2020

**Abstract:** This study advances the research and methodological approach to measuring and understanding national-level destination competitiveness, sustainability and governance, by creating a model that could be of use for both developing and developed destinations. The study gives a detailed overview of the research field of measuring destination competitiveness and sustainability. It also identifies major predictors of destination competitiveness and sustainability and thereby presents destination researchers and practitioners with a useful list of priority areas, both from a global perspective and from the perspective of other similar destinations. Finally, the study identifies two major types of destination governance with implications for research, policy and practice across the destination life-cycle. The research deals with the analysis of the secondary data from the World Economic Forum Travel and Tourism Index (WEF T&T). Major types of destination governance and predictors of belonging to either one of the types, as well as inside cluster predictors have been extracted through a two-step cluster analysis. The results support the notion that a meaningful model of national-level destination governance needs to take into account different development levels of different destinations. The main limitation of the study is its typology creation approach, as it inevitably leads to simplifications.

**Keywords:** innovation; indicators; governance; sustainability; competitiveness; destination; life-cycle

#### **1. Introduction**

A destination's success depends on its competitiveness in the global market, but also on the need to sustain its competitive position and be resilient in the face of unforeseen events as a prerequisite of long-range success [1,2]. This is a difficult task, because destinations are being produced and reproduced through a complex combination of social, cultural, political and economic relationships, making tourism research a transdisciplinary field of research, which beyond business research includes spatial issues (local, regional, national), thematic issues (mobility, culture, sustainability) and different approaches (advocacy, cautionary, adaptive and knowledge-based platforms) [3–5].

There is a gap in the literature on the most significant factors of destination performance that could be of use for both policy and organizations [6]. This exploratory study therefore seeks to fill this gap by creating a taxonomy model that could provide more flexibility in understanding the types of challenges faced by different destinations, and at the same time acknowledging that a global model of destination excellence needs to take into account a multitude of approaches to destination planning and development. For creating a taxonomy model, research deploys a two-step cluster analysis of the data from the two Travel and Tourism Competitiveness reports (Crotti and Misrahi [7] and Crotti and Misrahi [8]), thereby answering the call from Dwyer and Kim [9] for further research of data on competitiveness from the World Economic Forum.

Understanding major predictors of destination competitiveness is of essential importance for destination planning and governance arrangements. The importance of specific predictors (both in global terms and in terms of a narrower competitive set) is important for setting the agenda for discussions on the future destination planning and governance, and aligning the destination-level goals with the changes in the competitive set and in the global competitive landscape.

The article identifies major national-level destination competitiveness and governance types, predictors of belonging to either one of the types identified, as well as predictor importance inside each of the two competitiveness and governance types. Before presenting the results, a literature review summarizes previous articles on indicators of destination competitiveness and destination governance, while the discussion positions the results within the two research fields.

#### **2. Literature Review**

#### *2.1. Competition, Competitiveness*

Destination competitiveness is measured through specific competitiveness factors, especially focusing on specific factor sets that are of relevance in a specific destination competitiveness group and specific destination life-cycle stage [10–13]. Competition in general business terms is about success and about outperforming the others in a particular market by aligning one's firm's activities according to priorities and establishing a profitable and sustainable industry position [14,15]. However, regarding the competition between tourism destinations, it is a more complex phenomenon than inter-organizational competition for a number of reasons: (a) national tourist destinations belong to a specific (and non-changeable) competitive set because of geographic position, previous involvement with the global tourism industry and natural and cultural resources [16,17]; (b) there is a pronounced difference between inherited/endowed resources and created resources [9]; (c) the degree of (potential) tourist product complementarity determines the optimal level of competition or cooperation between regional destinations in the global market [18]; (d) major drivers of competitiveness are often non-economic, e.g., enhancing the well-being of destination residents or preserving natural resources [19,20]. The problem with applying the concept of competitiveness on national-level tourism destinations is that competitiveness is often viewed from the short-term perspective, particularly in times of crisis, to include strong promotional activities on international tourism markets, decreasing costs and identifying synergies between tourism actors [6]. An important distinction should be made at this point regarding comparative advantage (e.g., an abundance of natural resources, low labor costs) and competitive advantage (the ability to add value to the resources in order to sell them on the market) [21–24]. Competitive advantage represents the value that can be produced for the buyers that can exceed the cost of creating this value: value in this sense is what buyers are willing to pay [15]. Benchmarking is a tool often used for analyzing a destination's competitive position [25]. It can be conducted as internal, competitive, functional or generic, and it is an especially good tool for monitoring qualitative aspects of tourism development to systematically analyze performance, processes and strategies [26–30].

#### *2.2. Determinants of Destination Competitiveness and Sustainability along the Destination Life-Cycle*

Before making a more nuanced analysis of destination development, it is first important to understand what represents a successful tourism destination and what does not. This paragraph gives a short literature overview for indicators of destination and/or tourism performance. For a full list of major studies in this field please refer to Table 1 at the end of the literature review or consult the a review provided by Medina-Munoz et al. [31]. Assaf and Josiassen [6] identified the ten most negative and positive indicators of tourism performance. Taking into consideration these identified indicators, the goal for destination development would be to strive towards excellence to become, as Gilbert [32] defines it, "status areas" rather than "commodity areas", and attract high spenders and loyal tourists. The most significant obstacle in achieving this is that too extensive lists of destination competitiveness predictors lead to a need for determining the importance of each one of the predictors, as not each and every one can be of the same importance [10]. This is a major research gap identified in the literature that this article seeks to close, by providing a more usable set of most-relevant indicators for both developing and developed destinations.

The destination life-cycle model provides an argument that in developing destinations, demand should firstly exceed the supply, followed by a readjustment period in more mature phases, where high economic, social and environmental tourism impacts need to be managed [13]. However, later studies posit that tourism policy and decision making in developing countries needs to move away from putting a sole emphasis on quantitative measures of economic growth and enable qualitative measuring and destination development through better local stakeholder consultation early on in the destination development process [33–37]. In mature destinations, growth strategies are often connected to new product development that includes the expansion of: (1) networking between the actors, (2) customer value, and (3) competitiveness [38]. This is usually achieved by connecting destination resource space with activity space and experience space. Goffi and Cucculelli [39] single out in their research destinations of excellence (developed destinations) either based on their environmental standards (primarily related to water quality) or based on their built heritage and public services and activities, located within small, usually rural communities. In most rural destinations the emphasis is on creating tourism products related to natural resources, while in urban destinations, such as Dubai, the focus in on building global air accessibility as well as luxurious accommodation facilities [40,41].

Special attention should be given to emerging themes, such as Internet-related technology. In this sense, knowledge and innovation need to be the core value of tourism destination planning and development in order for the destination to survive in the global competitive environment [42,43]. The Internet and social media are one of the major megatrends having an impact on the society as a whole, and especially tourism, as a wide range of data is now available to tourists on the go: landscape descriptions, pricing, accommodation rating and local news [44]. Standing in relation to this aspect is the growing social importance of a digitally affluent generation, namely the millennials (generation Y), as they represent the future of both consumers and the job market, by including their vacation habits, sustainability attitudes, social media usage patterns, increasing participation in luxury markets and workplace preferences [45–47]. As consumers, millennials are often non-traditionalist in their choices even for luxury products [48,49].

Sustainability should play an important role in fostering long-term tourism destination competitiveness in developing destination [50–52], but it is even more important for the competitiveness of the developed destinations. One of the most important obstacles for implementation of sustainable tourism in developing destinations consists of managerial values and social representations of sustainability [53]. Regarding specific indicators, one of the major factors identified in the literature is air quality [54–56], especially in city destinations like Beijing, Dubai or Belgrade [41,57–59]. Other frequent environmental issues in destinations like Egypt, China, India, Montenegro, Croatia and Serbia include water pollution and inappropriate garbage disposal [58,60–62],

#### *2.3. Destination Planning, Development and Governance*

Destination governance encompasses both corporate and public governance and can mean both the architecture of relationships between public and private actors and the process of steering the society [63]. Angella, et al. [64] have extracted four types of destination governance: normative, leading firm, entrepreneurial and fragmented (scattered governance function, weak coordination mechanism). Major obstacles of national tourism destination governance include a complex and diffused action field and a limited reach when it comes to private actors at the destination [65]. Other

destination governance problems include: lack of, or inefficient, soft and interdisciplinary planning instruments; an insignificant role of destination residents in decision-making; a dominant role of foreign tour operators; and a power-distant government department and/or destination management organization [53,66,67]. In addition, DMOs (Destination Management Organizations) should be equipped with financial means and political and legislative power in order to be able to manage the interests, benefits and responsibilities of tourists, host population, tourism enterprises, tour operators and the public sector [13].

Successful destination governance needs to include common goals, a balanced power between the actors and co-evolutionary adaptations [68–71]. The phenomena related to poor governance mostly include hierarchical structures, lack of inclusion trust and perceived justice from actors [70,72], while the new and emerging theme in destination governance are public–private partnerships [73,74]. Nadalipour, et al. [75] call for future research on identifying a globally applicable model for investigation of tourist destinations in different contexts and their sustainability and competitiveness, by deploying multidisciplinary indicators of sustainable competitiveness. This research closes this research gap by acknowledging that a globally applicable destination governance model needs to be flexible enough to be used in different types of settings—both in terms of mutual relationships between major tourism actors as well as regarding processes steering tourism development. The reason for this is that different forms of multi-actor, networked collaboration arrangements directly impact the innovation of place-based competitiveness and sustainability policy [76–80]. This approach is becoming even more relevant in light of disruption caused by new technologies in the service industries: from tourism to hospitality and to mobility, new business models are disrupting business-as-usual and challenging the regulatory frameworks and the existing balance of power between the destination actors [81–84].

Sustainability is one of the most important concepts for the future of tourism governance [42,85]. However, as has been demonstrated in the literature, tourism has improved the socioeconomic conditions only in the most developed countries, while developing countries have problems with the implementation of sustainable tourism concepts because of pressurized political contexts: large-scale capital-intensive real estate projects are encouraged without having (or disregarding) an integrated plan to account for environmental and local community impacts [86–90].



#### **3. Methodology**

Although significant criticism of the World Economic Forum Travel and Tourism (WEF T&T) data and their mixed collection method has been presented in the literature, it is considered to be the most complete and relevant global data collection effort regarding destination competitiveness and sustainability, and as such suitable for further discussion of national-level tourism policy [17,39,97]. Therefore, data from the 2015 and 2017 WEF T&T reports [7,8] were used for this analysis. Data from previous reports (2008, 2009, 2011 and 2013) were excluded due to incompatible indicator selection due to considerably different methodology. The latest report (2019) was not yet available at the time of analysis. Firstly, the data from 2015 and 2017 were cleaned to include a consistent set of countries (131) and variables and indicators (86). A total of 10 countries and variables were deleted because they were not present in both reports, as well as two indicators that had missing values. For an additional four out of 86 indicators used, the data were present only in one of the two reports, and thus no average was calculated for these four indicators. For the remaining 82 indicators, the average of the indicator values from both reports (2015 and 2017) was calculated. For this data set, a two-step cluster analysis was calculated using IBM SPSS 23 software.

Regarding cluster quality in terms of their cohesion and separation, the average silhouette value was 0.5, pointing to a good fit both by SPSS green color indication and as confirmed in the literature by Sarstedt and Mooi [98]. This was achieved by choosing a solution with 23 inputs and two clusters. There were four other solutions that reached the 0.5 silhouette value, all including the two-cluster solution, but with a higher number of inputs (25, 28, 31 and 34). The solution with two clusters and 23 inputs was therefore deemed the most compact and useful model in this group. By deploying this procedure, answers to the following research questions were sought:


#### **4. Results**

By deploying a two-step analysis, two major types of destinations were extracted—developed ones (scoring higher on all relevant 23 indicators on average) and less developed ones (scoring lower on all relevant 23 indicators), as presented in Table 2. Firstly, the overall indicator relevance for the clustering solution was shown (in descending order), where indicators are called predictors. In order to further delve into the specificities of both clusters, in Table 3, the 23 indicators were presented according to their inside-cluster importance, in descending order.

The following predictors were used for the two-cluster solution: Wastewater treatment (1.00); Fixed broadband Internet subscriptions (0.81); Ground transport efficiency (0.80); Quality of roads (0.78); Quality of railroad infrastructure (0.75); Reliability of police services (0.72); Ease of finding skilled employees (0.69); Degree of customer orientation (0.68); Internet users (0.67); Quality of air transport infrastructure (0.66); Enforcement of environmental regulations (0.66); Paved road density (0.62); Mobile-broadband subscriptions (0.60); Quality of electricity supply (0.59); Quality of port infrastructure (0.57); Purchasing power parity (0.53); Number of international associations meetings (0.53); Number of operating airlines (0.46); Aircraft departures (0.45); Cultural and entertainment tourism digital demand (0.42); Pay and productivity (0.41); Stringency of environmental regulations (0.39); and Available seat kilometers, international (0.35).


**Table 2.** Extracted clusters and major predictors in the two-step cluster analysis.

**Table 3.** Cluster size and major inside-cluster predictors extracted in the two-step cluster analysis.


The two created clusters are presented in Table 3 (developed vs. developing), with the accompanying in-cluster importance of specific predictors. These results lay a foundation for differentiated destination governance theories for developing and developed destinations.

#### **5. Discussion, Limitations and Future Research Directions**

Tourist destinations are often being compared regarding the number of overnight stays or tourist arrivals, the share of overnight stays or tourist arrivals in a specific market, or corresponding growth rates—an approach based on a classical TALC (Tourism Area Life Cycle) model of destination development [12,99]. However, dealing only with the number of tourist or overnights has its disadvantages, as it does not take into account prices or quality attributes [95,100]. More importantly, recent research has demonstrated the unreliability of official statistics due to manipulation of taxable overnight stays by accommodation providers [101]. This article fills this research gap by contributing to the existing knowledge on destinations competitiveness and sustainability, by providing benchmarking and indicator weighing for both developing and developed destinations. Two major types of destinations (developing and developed) were extracted, as well as one overall and two type-dependent predictor lists that enable better understanding of the global destination competitiveness.

The research results confirm the findings from the literature [19], that different destination competitiveness factors (predictors) have different impacts on the competitiveness of developing and developed destinations, going even further to rank the factors according to their relevance for both types of destinations. Identified predictors of global destination excellence, as well as inside-cluster relevance for both groups, should be further investigated and used for creating weighting schemes for indicator systems in different destinations. In other words, different indicators should be weighed in accordance with their importance (from 1.00 to 0.35), thereby closing to a big extent the research gap on weighing schemes, as identified by Zehrer, Smeral and Hallmann [95].

Having in mind the high relevance of the Internet-related indicators (Numbers 2, 9 and 13), more attention should be given to the Internet, social media, and how the digitally-oriented millennial generation is changing destinations globally—as consumers, as a workforce and as citizens in both developing and developed destinations.

The high importance of sustainability for tourism destination competitiveness on the global level, and especially for developed destinations, has been confirmed in the research. The results should serve as a starting point for tackling attitude–behavior gaps of destination managers and other stakeholders regarding sustainability. The environmental aspects captured by the model are: (1) Wastewater treatment, (11) Enforcement of environmental regulations, and (22) Stringency of environmental regulations. There is also a big difference in the municipal waste management and generally circular economy capabilities between developed and developing countries, which can all negatively affect the tourism industry in developing countries.

The research results emphasize the importance of a stable electricity supply and Internet use in developed destinations, coupled with physical infrastructure development, degree of customer orientation and workforce training and development, as well as reliable police services and wastewater treatment.

The research findings tackle the practical, managerial side, by extending the approach already deployed in the literature [93] and providing an alternative framework to be used on the national, regional or micro scale for accessing and weighing the competitiveness and sustainability of a destination in the global context. The findings also enhance the value of the Travel and Tourism Competitiveness Index, by making it more approachable for destination managers. The results also provide empirical evidence that quantitative growth in developing destinations (in this case of air transport traffic, purchasing power parity and international association meetings) needs to go hand in hand with wastewater treatment improvement and stringent environmental regulation, coupled with further digital and physical infrastructure development, as well as workforce training and development. There are also further considerations to be dealt with in politically unstable destinations (such as the island of Cyprus), where regional visitation is highly dependent on the perceptions of culture and ethnicity [102]. Similarly, post-war destinations face highly specific tourism development problems, such as lack of basic political prerequisites for the functioning of society, while the need for active re-branding and infrastructure re-development seems to be a top priority [103–108].

Considering the St. Gallen Model of Destination Management, being focused on developed destinations, it constantly redefines and updates the definition of a destination, and also discusses the DMO's role in a destination-level network, as well as destination leadership, strategy, resilience and governance arrangements [109,110]. However, the two-step clustering solution presented in the results section confirms the findings of previous studies, that there are significant differences in the process and outcomes of tourism development in the developed and developing countries [90,111]. Therefore, both types of destinations are presented in Figure 1, so as to better visualize the tourism destination governance arrangements and their mutual differences. The model builds on the premise that destinations first need to be in the type 1 destination governance mode in order to advance to the type 2 destination governance mode at a later point in time.

**Figure 1.** Destination governance typology.

The research results complement and extend quantitative measures of destination competitiveness, related to tourist numbers and GDP, which appear to still be relevant in many developing destinations. However, these quantitative measures need to go hand in hand with social and environmental indicators [112]. Therefore, the optimal way of measuring global destination competitiveness is by deploying a model that makes a distinction between developing and developed destinations, each with their set of destination governance priorities. However, global destination governance priorities (common to both destination governance types) are also being identified, and can be seen as long-term and basic priorities for both types of destinations, while in-cluster priorities have more relevance for each competitive set. Therefore, governance types are mainly understood here as stakeholder importance and the consequent power relationship architecture between different types of actors at the destination. The model does not consider governance arrangements or processes in either developed or developing destinations. Future research can investigate the precise inside cluster weighing of predictors, in order to develop a weighing scheme for both developing and developed destinations.

Identified predictors of global destination excellence should provide stakeholders in both developing and developed destinations with an early discussion basis for anticipating change and making timely destination governance arrangements and adopting a long-term global perspective, regardless of the current level of development. Going a step deeper into the in-cluster predictors, destinations can decide on a set of governance priorities of more direct relevance to competitiveness inside one's own competitive set.

#### **6. Conclusions**

The article started by giving an overview of the literature on destination competitiveness, the predictors of destination competitiveness and sustainability and of destination planning, development and governance. It then presented an exhaustive overview of approaches to measuring destination competitiveness and sustainability—from the number of indicators used (observable variables), concepts used to classify the indicators (non-observable mediating variables) and methodology used to the analysis of the data collected. There are both inductive and deductive approaches in this research field, but the main weakness in inductive approaches seems to be the creation of one single model of destination competitiveness to be applied to all destinations, usually by applying PCA (principal component analysis). This statistical method is rather a dimension reduction method than a proper clustering method. In order to fill the research gap and answer the first research question, this research deployed a novel method—a two-step cluster analysis—and identified two major global types of destination competitiveness—one for more developed destinations and the other for less developed destinations. The created model gives a comprehendible list of major predictors for belonging to either one of the two competitive sets, thereby answering the second research question. The model also provides a within-cluster importance rank for both competitive sets, thereby answering the third research question. In this way, a very usable and action-oriented model was created for both academicians and destination managers to be used in further research globally. The identified predictors can provide the most important factors of moving the destination from a lower-level development to a higher-level development. In practice, this would usually mean either development or consolidation for an already developed destination that has experienced a downturn.

The major limitation of this study relates to the methodological problems when attempting to aggregate large amounts of data from different fields of society. The second limitation relates to the induced model with two major types of destination competitiveness and sustainability, as it is inevitably a logical simplification of the reality of global destinations. Although it can be useful for starting a discussion on major types of global destination competitiveness, sustainability and governance arrangements, it is still far from identifying all boundary conditions and outcomes of successful destination development. Another important issue is that some important indicators from the literature (e.g., air quality) are not included in this list, but have been demonstrated to be of great importance in many destinations. This is why contingencies regarding the application of the model in different regional, national or local contexts should be further identified and analyzed with the help of other research methods.

The major goal of the study was to contribute to the literature on destination governance, by deploying a novel method for creating a destination typology based on stakeholder prioritization by extracting major predictors of belonging to each one of the two types: developed and developing destinations. Further research should concentrate on extracting further specific governance types according to specific geographic areas, narrower competitive sets and other aspects of destination governance, beyond stakeholders—power relations, governance structures or processes.

This novel methodological analysis approach to destination competitiveness strengthens the indicator-driven policy analysis by creating a reference model with two different destination types. This is of relevance for both academics as well as practitioners. Furthermore, the results demonstrate the importance of making a distinction between developed and developing destinations when considering different competitiveness and sustainability models. The results also enable the creation of weighing schemes to more precisely measure destination competitiveness and sustainability in different contexts.

**Author Contributions:** Conceptualization, I.P.; Methodology, M.D. and I.P.; Software, I.P.; Validation, T.M.N. and S.P.P.; Writing—original draft preparation, I.P., T.M.N. and S.P.P.; Writing—review and editing, M.D.; Supervision, M.D.; All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

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

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Structure and Formation Mechanism of China-ASEAN Tourism Cooperation**

**Jie Yin 1, Yahua Bi 2,\* and Yingchao Ji <sup>1</sup>**


Received: 16 June 2020; Accepted: 2 July 2020; Published: 6 July 2020

**Abstract:** Tourism cooperation is an essential element for tourism development in China-ASEAN countries and has made a significant economic contribution to destinations. This study investigates the structure of tourism cooperation in China-ASEAN relations and identifies a set of factors that affect tourism cooperation from a network perspective. By employing social network analysis, the results indicate that the scale of cooperation is small, and the efficiency is not high, although the restrictions on cooperation between countries are reduced. The findings also indicate that differences in the political system, security, population density, and language can promote tourism cooperation, while differences in governance, income, and consumption level impede tourism cooperation. The research results may assist China-ASEAN countries to formulate tourism strategies suitable for international cooperation and national differences.

**Keywords:** tourism cooperation; China-ASEAN; cooperation structure; driving factors; regional tourism

#### **1. Introduction**

With the in-depth development of tourism, competition in the tourism market is becoming increasingly fierce [1]. Under the complex and competitive atmosphere, tourism cooperation, as an important element for tourism destinations to obtain competitiveness [2], has become a vital consideration for practitioners and scholars. Morrison et al. [3] pinpointed that various countries utilize partnerships to develop tourism, indicating that tourism cooperation becomes an organization's preference [4,5]. As such, tourism cooperation is regarded as an important approach to promote the sustainable development of tourism [6]. Researchers have examined the issues surrounding tourism cooperation in various industries [7–9], such as sport industries [10] and forest, mining, and tourism industries [11]. In the context of tourism, scholars mentioned that tourism cooperation can be explored through the cooperation network [5].

The cooperation network is identified as a coherent pattern of interactions and interconnections between organizations, as opposed to such organizations being isolated in the system [5,12]. In particular, through the cooperation network, organizations collaborate to obtain mutual benefits and win-win results [9,13]. The network, as a concept, has been widely adopted in international tourism [9,14].

However, although cooperation projects related to international tourism have been launched globally [15,16], international tourism cooperation networks of these projects have been ignored to a certain extent [3]. It is unclear how international tourism cooperation projects interact, especially from a perspective of the network structure. Concerning these successful tourism cooperation projects, it is necessary to interpret the characteristics of tourism cooperation and the factors that influence tourism cooperation.

The Association of Southeast Asian Nations (ASEAN, including Brunei, Cambodia, Indonesia, Laos, Malaysia, Myanmar, the Philippines, Singapore, Thailand, and Vietnam), a well-known cooperation project, is experiencing a boom in both foreign and domestic tourism with tourism becoming one of its foremost industries [17]. Tourism is one of the main priority sectors for ASEAN [18]. Furthermore, the ASEAN National Tourism Organizations (ASEAN NTOs) formulated a plan of action for ASEAN cooperation in tourism. The issues for the tourism cooperation of ASEAN have drawn increasing attention. Issues of tourism demand [19–21], tourism flows [22], cooperation trends and prospects [23], intergovernmental collaboration [24], supranationalist alliances [25] and the preconditions and policy framework [26] of tourism cooperation in ASEAN have been done in the past. Additionally, as an important partner and neighboring country of ASEAN, since 2012, China has become the source of the largest outbound tourism globally [27] and one of the world's major recipients of tourists [28] with the number of inbound visitors expanding enormously. China has conducted cooperation with ASEAN in different fields, such as trade [29,30], security [31], economic [32] education [33] and tourism [18,22,34–36].

What is more, for ASEAN countries, the most important goal is to maintain close and beneficial cooperation with each other [19], especially tourism cooperation that brings huge economic benefits. Beritelli [2] suggested that it is effective to identify the codependent relationship between partners from the network perspective to induce cooperative behaviors. However, given the research on the tourism cooperation of ASEAN, relatively little research has explored how tourism cooperation is formed and what are the characteristics of tourism cooperation. Within this context, strengthening tourism cooperation with China can achieve close cooperation in China-ASEAN relations. Therefore, this research complements previous research on China-ASEAN by exploring the structure of their tourism cooperation with its characteristics and relationships, and investigating the driving factors and formation mechanism while deconstructing the tourism cooperation from the network perspective.

To accomplish it goals, this research adopts social network analysis (SNA) to examine tourism cooperation. First, this study elucidated the cooperation structure while exploring the essential characteristics of China-ASEAN tourism cooperation by taking overall characteristics and individual characteristics into account from 1998 to 2017 (The reason for choosing this period is that various events affect international tourism cooperation during this period. In 1967, ASEAN was formally established. ASEAN membership reached nine countries in 1997 and ten countries in 1999. In 2001, China joined the world trade organization. SARS broke out in 2003, the global financial crisis broke out in 2009, and the Belt and Road Initiative (BRI) was put forward in 2013. To comprehensively analyze the structure, characteristics and formation factors of China-ASEAN tourism cooperation, the years before and after the formation of the 10 countries as well as the years that affect tourism cooperation by international events were included in the research period). Second, this study explained the reasons that such a characteristic cooperative network is formed by investigating what the essential driving factors are affecting the tourism cooperation network (The driving factors affecting tourism cooperation networks would be explored by employing Quadratic Assignment Procedure (QAP) analysis, which was introduced in detail in the following context).

In summary, the China-ASEAN relations in the current study provide an excellent research context for understanding tourism cooperation. This study also provides two important contributions to the tourism literature: (1) This study shows the essential factors that affect tourism cooperation. In detail, differences in terms of income, governance, and consumption level have negative effects on cooperation networks presently, whereas differences in population density, security, and the political system promote cooperation networks. The findings enhance the comprehension of tourism cooperation in the China-ASEAN network. (2) This study clarifies the structure of tourism cooperation in China-ASEAN and provides valuable references for increasing tourism benefits and establishing new strategic plans. Overall, figuring out the structure and examining the influencing factors of tourism cooperation in China-ASEAN countries helps remove cooperation obstacles, optimize cooperation structure, and promote sustainable tourism cooperation among China-ASEAN countries. Therefore,

the findings of this research will contribute to a deeper and more valuable understanding of international tourism cooperation.

#### **2. Literature Review**

#### *2.1. Tourism Cooperation Network*

Tourism cooperation enhances regional relations and drives regional economic development; as such, tourism cooperation is regarded as an effective way for the sustainable development of tourism [5,37,38]. Considering the vital role of tourism cooperation, past research has investigated tourism cooperation on travel behaviors [39], tourism establishments [5], and tourist movement patterns [40,41].

Tourism development is accompanied by fierce competition [42]; therefore, scholars emphasized that the issues related to tourism cooperation need to be addressed [43–45]. Early research on tourism cooperation mainly focused on cooperation obstacles, opportunities, strategies, and methods [34,35,43,45,46]. Jamal and Getz [43] proposed the principles for urban tourism cooperation from a planning view. Cetinski and Weber [45] explored the possibility of establishing sound cooperation among the multinational tourism markets. Elliott [47] analyzed the measures on cooperative management among administrations. Bramwell and Angela [44] proposed a framework for tourism cooperation decision making.

Additionally, the advancements of transportation and globalization have promoted tourist flows, thus forming a diversified tourism phenomenon. Then, the network theory and network science approaches [48,49] were introduced into tourism to reveal the complex tourism phenomenon. According to network theory and network science, a network describes organizations aligning together to form inter-organizational networks or a type of flexibly designed network structure [50]. In other words, the network refers to a special structure consisting of different actors or organizations and their connections with others. Chung et al. [51] employed the social network analysis approach to reveal that global tourism networks have become highly consolidated. Provenzano and Baggio [52] also found that Sicily has a complex destination structure through inbound tourism. Yi et al. [53] observed that the networks in village tourism committees in China are diffuse. Due to the advantages of revealing connections and structure, the network science approach has been widely employed in tourism cooperation.

The network has been adopted in various tourism studies [9]. Past studies have been interested in tourism destination cooperation [54], tourism enterprise cooperation [55], and tourism geography cooperation [56]. Tourists move on a worldwide scale currently and construct a heterogeneous and complicated network [57]. However, little research focuses on international tourism cooperation, indicating that international tourism cooperation networks are still a relatively neglected area [3]. To bridge the research gap, this study investigates cross-border tourism cooperation from the network perspective.

#### *2.2. Research Context: China-ASEAN*

Even though tourism cooperation in the China-ASEAN counties has become close and frequent, academic studies related to intergovernmental collaboration in tourism among China-ASEAN counties seemingly remain few in number [26]. Chirathivant [23] discussed the trends and prospects of ASEAN-India tourism cooperation. Chang, Khamkaew, Tansuchat and McAleer [20] applied a multivariate conditional volatility model to investigate the interdependence of international tourism demand and encouraged regional cooperation in tourism development among ASEAN member countries. In recent years, tourism cooperation among ASEAN countries has made some progress. For Thailand, the number of tourists to Thailand from ASEAN countries and Thailand's foreign exchange earnings saw an average growth rate of more than 10% [58].

Despite ASEAN countries having numerous opportunities for tourism cooperation, they still face various challenges [24,59]. Koh and Kwok [59] assessed the progress undertaken by the ASEAN establishment in terms of tourism development and evaluated the possible challenges of such intra-regional cooperation. To promote tourism cooperation, Wong, Mistilis and Dwyer [24] explored the factors that facilitated and hindered progress for tourism cooperation in ASEAN and identified that the lack of implementation of tourism integration has hindered the promotion of tourism cooperation. Then researchers explained the preconditions that gave rise to ASEAN tourism and the formulation of the policy framework [26]. Based on the promotion factors, obstacles, and prerequisites of tourism cooperation, the mechanism of ASEAN tourism collaboration [60] were displayed.

Given the lack of analysis of tourism cooperation from an empirical perspective, it is crucial to understand the tourism cooperation relationships between China-ASEAN counties. More importantly, China-ASEAN countries need to promote the sustainable growth of tourism [61]. Thus, this paper focuses on the tourism cooperation of China-ASEAN and intends to reveal the characteristics of tourism cooperation.

#### *2.3. Determinants of Tourism Cooperation in China-ASEAN and Hypothesis*

Even though the previous studies investigated certain issues on tourism cooperation of China-ASEAN, there seems to be a lack of experimental testing of the determinants of collaboration. Thus, this paper makes efforts to empirically examine the determinants of tourism cooperation. Concerning the factors affecting international tourism cooperation, Czernek [8] proposed some main issues, including exogenous factors (i.e., economic, income, political changes), endogenous factors (i.e., cost, level of tourism development, geographical distance, and political changes), and global factors (i.e., global environment). Wong, Mistilis and Dwyer [26] argued that political, social, and economic development alongside variations may be the preconditions for tourism cooperation of China-ASEAN. Curiosity is one of the main motivations of tourists [62], which means the difference between the source country of tourists and the destination country is an important reason for tourists to form international tourism activities. Additionally, the premise of cooperation is that partners can complement each other [32]. Based on the aforementioned discussion, we argue that the difference may be the basis of tourism cooperation. Therefore, according to the influencing factors of tourism cooperation, we aim to investigate and empirically test how these factors affect the tourism cooperation in China-ASEAN from the perspective of differences.

(1) Political system difference (PSD): Numerous studies have pointed out that political factors can influence the development of tourism [63–66]. For ASEAN countries, political factors have greatly affected the development of tourism [67–69]. The political system is the comprehensive embodiment of political factors and the cohesion of a country's political factors. Differences in the political system could lead to great differences in policies and development directions, which may make cooperation between the two countries difficult to occur. Based on the above analysis, the hypothesis was posited below:

#### **Hypothesis 1.** *The di*ff*erence in the political system negatively a*ff*ects tourism cooperation in China-ASEAN.*

(2) Governance difference (GD): Governance is a key concept in politics and public policy [70], which reflects the government's comprehensive measures on a series of issues including industry, society, and livelihood. For tourism, governance capacity is an important guarantee for tourism development [71,72]. If the governance of a country is high, there is a positive environmental benefit for its tourism industry. Therefore, we argue that the great difference in governance between the two countries may result in a huge difference in the environment for the tourism industry. The difference in the tourism industry development may restrict the complementary advantages of limiting tourism cooperation. Accordingly, the relationship between governance difference and tourism cooperation is postulated as follows:

#### **Hypothesis 2.** *The di*ff*erence in governance negatively a*ff*ects tourism cooperation in China-ASEAN.*

(3) Income difference (ID): Tourism activities need the support of discretionary income [35]. Moreover, income represents a country's economic development level to some extent. As for international tourism activities, tourist flows are affected by numerous economic factors, such as price, income, and exchange rate [36,73–76]. In general, most countries may be reluctant to cooperate with countries with weak economies. Hence, in this situation, tourism cooperation between the two countries may be weak. If there is a big income difference between countries, tourist flows between them are difficult. Thus, we propose that:

#### **Hypothesis 3.** *Income di*ff*erence negatively a*ff*ects tourism cooperation in China-ASEAN.*

(4) Consumption level difference (CLD): In essence, tourism is a consumption activity, which is largely influenced by the consumption level of destination [77]. The higher the consumption level in the destination country is, the higher the cost for foreign tourists to travel in the country. However, the travel cost would affect the demand for travel [78]. Therefore, we claim that the difference in consumption level may restrict tourists' demand, which is negative for tourism cooperation. Based on the above discussion, it is hypothesized that:

#### **Hypothesis 4.** *Consumption level di*ff*erence negatively a*ff*ects tourism cooperation in China-ASEAN.*

(5) Security difference (SD): A safe destination environment is an important basis for tourists to travel to that destination [79]. Generally, tourists would not go to unsafe destinations. For tourism cooperation, crime rates are important problems that ASEAN countries currently face in the tourism industry [19]. Therefore, it is suggested that tourists travel to countries with low crime rates. This means that the greater the security difference, the more likely tourists travel, which leads to the following hypothesis:

#### **Hypothesis 5.** *Security di*ff*erence positively a*ff*ects tourism cooperation in China-ASEAN.*

(6) Population density difference (PDD): Anser et al. [80] observed that population density substantially decreased inbound tourism and international tourism receipts. Currie and Falconer [81] claimed that low population density benefits tourism development. High population density limits tourism demand. Therefore, tourists from a country with a higher density in the population are likely to travel abroad to hunt for a low-population-density country. Accordingly, we infer that:

#### **Hypothesis 6.** *Population density di*ff*erence positively a*ff*ects tourism cooperation in China-ASEAN.*

(7) Language difference (LD): Concerning the international tourism, language is treated as the cost for tourists [51] as languages are different among countries and regions. Thus, the difference in language is included in constraining factors for international tourism [82]. As for convenience, tourists would select the destination country speaking the same language. As such, the following hypothesis is proposed:

**Hypothesis 7.** *Language di*ff*erence negatively a*ff*ects tourism cooperation in China-ASEAN.*

#### **3. Research Design**

#### *3.1. Measurement for Tourism Cooperation*

Through the relationships between social networks, different connections can be understood, such as communication connections and network relationships [83]. In the 1960s, the gravity model was

introduced [84,85] and then was widely employed to measure the relations in the tourism cooperation network [37,53]. Moreover, various studies utilized the gravity approach to explore international tourism dynamics in the tourism network [51], which denotes that the model is well applied to interpret the cooperation network in the global tourism environment [51,86]. Hence, the current research selected the gravity model to explore tourism cooperation connections in China-ASEAN counties. Based on Equation (1) of the gravity model, tourism cooperation connections were examined.

$$F\_{ij} = \sqrt{T\_i I\_i} \* \frac{\sqrt{T\_j I\_j}}{D\_{ij} \* D\_{ij}} \tag{1}$$

where *Fij* represents the tourism cooperation links between country *i* and country *j*; *Ti* and *Tj* are the number of tourists in country *i* and country *j*, respectively; *Ii* and *Ij* are the tourism income of country *i* and country *j*, respectively; and *Dij* is the geographical distance between country *i* and country *j*.

However, spatial distance [51], time distance [87], and cultural distance [88], etc., can influence the intensity of tourism cooperation to a certain extent. The aspects that affect tourism cooperation make it necessary to modify the gravity model instead of using a single gravity model. Therefore, researchers recommend using a modified gravity model to interpret the tourism cooperation connection [89,90]. Additionally, tourism cooperation has strong economic characteristics, such as economic distance, affect tourism relations between countries [91]. In the tourism environment, tourism cooperation could be also influenced by the industrial development environment, such as local culture and service quality [16,92–94]. Based on the aforementioned discussion, the current research introduces economic distance and the industrial development environment to modify the gravity model (see Equation (2)).

$$F\_{ij} = K\_{ij} \ast \sqrt{T\_i I\_i} \ast \frac{\sqrt{T\_j I\_j}}{G D\_{ij} \ast E D\_{ij}} \tag{2}$$

where *Fij* refers to the tourism cooperation links between country *i* and country *j*; *Kij* represents the industrial development environment, describing the tourism cooperative attraction coefficient of country *i* and country *j*. *Kij* is acquired through Equation (3) [95]. Additionally, *GDij* and *EDij* represent the geographical distance and economic distance between country *i* and country *j*, respectively.

$$K\_{ij} = \frac{SI\_i}{SI\_i + SI\_j} \tag{3}$$

where *SIi* and *SIj* are the ratios of employment in the service industry to the total employment for country *i* and country *j* (i.e., the proportion of employment of the service industry in total employment). By taking the economic distance measurement method [15], economic distance was investigated by Equation (4).

$$ED\_{ij} = \frac{\left(GDPPC\_i - GDPPC\_j\right)^2}{GDP\_i \* GDP\_j} \tag{4}$$

where *EDij* is economic distance between country *i* and country *j*; *GDPPCi* and *GDPPCj* are the per capita GDP of country i and country j, respectively; and *GDPi* and *GDPj* are the GDPs of country *i* and country *j*, respectively.

#### *3.2. Social Network Analysis*

Social network analysis interprets social cooperation through the network [96–98] and is applied in various tourism studies [40,99,100]. For example, Leung, Wang, Wu, Bai, Stahura and Xie [40] employed the social network to investigate tourist movement patterns. In the study of Luo and Zhong [100], communication characteristics of word-of-mouth in tourism interaction were explored by adopting the network analysis. The method is well developed to systematically study the social structure by measuring network density, centralization, betweenness, and structural holes [98]. Besides, the Quadratic Assignment Procedure (QAP) is commonly employed in the social network to investigate the influencing factors of the network [101,102].

Network density reflects the ratio between the actual link and the maximum number of links in the network [39]. A high density indicates a tight network connection.

Centralization contains the degree, betweenness, and closeness of centralization. When the centralization is close to 1, the network is close to concentration [103].

Betweenness is mainly used to measure the ability of individuals in the network to act as "intermediaries" and "mediators", which represent the individual's "control ability" [104]. The high betweenness indicates the importance of master resources and information flow, and the lack of such betweenness can cause communication failure to other connections as well.

Structural holes are explored by effective size, efficiency, and constraints [105]. The effective size measures the control power of a node in the network. The larger the value, the stronger the control power of the node. Efficiency reflects the degree of influence of a node on other nodes in the network. The larger the value is, the stronger the influence. Constraints denote the degree of a node to utilize structural holes. The smaller the value, the higher the degree is.

QAP (Quadratic Assignment Procedure) regression analysis was employed to understand the factors influencing tourism cooperation. The QAP regression analysis can efficiently reduce multicollinearity issues [106,107]. The correlation coefficient can be obtained by the QAP regression analysis using permutation matrix data, and nonparametric tests were then operated on the matrices to discover major aspects that influence tourism cooperation.

#### *3.3. Data Collection*

Tourism cooperation in China-ASEAN countries aims to increase the number of tourists and tourism revenue. Hence, this study uses the number of international tourists and international tourism expenses (Equation (2)) to measure cooperative ties between countries. The proportion of the employment of the service industry in total employment (i.e., the proportion of employment of the service industry in total employment) was applied to calculate the industrial development environment [95].

Concerning the determinants of cooperation, if two countries have the same political system, there is no difference in the polity between the two countries, and the political system difference (PSD) is 0; otherwise, it is 1. Six indicators ("voice and accountability, political stability and absence of violence/terrorism, government effectiveness, regulatory quality, rule of law, and control of corruption") [108] were regarded as the worldwide governance indices. The current research applies the average of the above six indicators to test the governance, which has been well utilized in previous studies (e.g., [109,110]). For language difference (LD), when two countries use the same official language, the value is set to 1; otherwise, set the value is set to 0. The income of tourists is measured by gross national income (GNI) and income difference is measured by the gap of GNI of two countries. The consumption level difference (CLD) is measured by the gap in per capita consumption of two countries. Security difference (SD) between the two countries is computed based on the difference in intentional homicide rates obtained from the source of the United Nations Office on Drugs and Crime's International Homicide Statistics (UNODC). Population density difference (PDD) is measured by the gap in population density between the two countries.

To test the indictors' differences among countries including GD, ID, CLD, SD, and PDD, we measured the subtraction of the indictor among China-ASEAN countries year by year and used the average of these subtractions as an index for identifying the indictor's difference. In detail, if the subtraction of the indictor of the two countries is less than the index, we then argue that there is no difference in the indictor between the two countries and set it as 0. If it is greater than the index, we suggest that there is a difference by setting it as 1.

The data (number of international tourists (T), international tourism income (I), GDP, GDP per capita, service industry employment, total employment, GNI, population density, per capita consumption and governance) calculated in the current research is derived from the World Bank (https: //data.worldbank.org). The official language data and spatial distance data (the spatial distance from one capital to another is regarded as the geographical distance between the two countries in China-ASEAN) use the database generated by CEPII (http://www.cepii.fr/CEPII/en/bdd\_modele/bdd.asp). In this paper, the data range is from 1998 to 2017.

#### **4. Results**

#### *4.1. The Structure of Tourism Cooperation in China-ASEAN*

#### 4.1.1. Characteristics of the Tourism Cooperation Network

Based on the revised gravity model, we calculated the tourism cooperation links among China-ASEAN countries from 1998 to 2017. These cooperation links among countries are employed to construct tourism cooperation networks among China-ASEAN countries with the help of social network analysis. In this study, we set the average value of the tourism cooperation links from 1998 to 2017 as the breakpoint value. Then, we set the relationship value as 1 if the value of the tourism cooperation relationship between the two countries was higher than the average value; otherwise, it was set to 0. We construct the tourism cooperation relationship matrix, and then show the visual analysis for the tourism cooperation network based on the closeness of each country for five stages, including 1998, 2002, 2007, 2012, and 2017 (see Figure 1). The closeness of a country means the extent to which it is not controlled by other countries in a cooperative network. Generally, the higher the value of the closeness is, the freer of the country is. As for calculating the number of the relationship value in these five charts, the numbers of cooperation ties in China-ASEAN countries are 39 in 1998, 38 in 2002, 43 in 2007, 50 in 2012 and 56 in 2017, indicating that the linkages of tourism cooperation networks are more intensified over time as revealed by the number of cooperation relations that have increased stably. According to Figure 1, we find that Singapore and Brunei are largely on the periphery of the tourism cooperation network, with few links to other countries.

**Figure 1.** Tourism cooperation networks on China-ASEAN in 1998, 2002, 2007, 2012 and 2017.

Additionally, this study uses the network density centralization to measure the overall characteristics of tourism cooperation networks of China-ASEAN countries. The results are shown in Figure 2.

**Figure 2.** Overall characteristics of the tourism cooperation networks from 1998 to 2017.

Network Density: The density of the tourism cooperation network grew stably from 1998 to 2017, as shown by the density value increases from 0.3545 in 1998 to 0.5091 in 2017. However, we revealed that there was a temporary decline in 2002, 2010, and 2011. In 2001, In 2001, the 9/11 incident occurred, which resulted in a sharp downturn in the international tourism industry [111] by presenting a decline in tourism cooperation in China-ASEAN countries. The international financial crisis in 2009 has had a huge impact on the world economy, which may have resulted in the decline of tourism cooperation in 2010 and 2011.

Centralization: The degree of centralization of the network has a fluctuating growth trend from 1998 to 2017, but it does not exceed 40%, indicating that there is not an obvious leader shown in this network. However, the betweenness centralization shows a fluctuating downward trend over the data period. It represents that the role of "intermediary" might not be as important as before. We argue that this essential finding might result from the role of direct cooperation being enhanced among countries, which might weaken the role of "intermediary" in this network.

#### 4.1.2. Individual Characteristics of the Network

By analyzing the betweenness and structural hole of the tourism cooperation network, we can derive the change of individual characteristics, including betweenness, effective size, efficiency, and constraints for the cooperation network and the results are shown in Table 1.

Concerning the betweenness, before 2013, the degree of betweenness presents a significant decline, which means that the role of "intermediary" for the China-ASEAN tourism cooperation network was weakened, as revealed by the maximum value decrease from 17.778 (China) in 1998 to 6.44 (Indonesia) in 2013 in Table 1. However, after 2013, the betweenness presents an increase. Furthermore, China acted as the middleman for the entire cooperation network. In 2013, the Belt and Road Initiative (BRI) was proposed by China. China has acted as the initiator and main promoter of the BRI, which is probably the main reason why it has become the intermediary for the China-ASEAN tourism cooperation network.

As for the change of effective size,Table 1 shows that the effective size of the country ranking first has not changed significantly (fluctuating around 3). Thus, we argue that the effective size of the China-ASEAN tourism cooperation network is still relatively small, and the scale of tourism cooperation needs to be further expanded presently. Besides, leading countries of effective size are constantly changing, which means that there are no strong leaders in the network.


**Table 1.** Ranking the first country for diverse individual characteristics from 1998 to 2017.

Regarding changes inefficiency, we revealed that the country that plays the most powerful role in the network is often changing since Thailand, China, Indonesia, and the Philippines have taken up this role successively. However, we also found that the overall efficiency of the network is decreasing because some countries do not have as powerful an influence on the cooperation networks of these countries as before, implying that cooperation network is moving towards the direction of equality and balance among these countries.

About the change in constraints, we noticed that the degree of constraints for China-ASEAN countries is gradually decreasing, indicating that cooperation liberalization and facilitation are gradually improving. As a result, we suggest that countries with lower constraints such as China, Indonesia, Vietnam, and the Philippines take the initiative to cooperate with other countries.

#### *4.2. The Formation Mechanism of Tourism Cooperation of China-ASEAN*

By employing the QAP regression approach, we explored whether tourism cooperation network would be affected by these factors including political system difference (PSD), governance difference (GD), income difference (ID), consumption level difference (CLD), security difference (SD), population density difference (PDD), and language difference (LD) for China and ASEAN countries. The results are presented in Table 2.

Table 2 reveals that political system difference has a positive effect on tourism cooperation. In particular, it has significant effects on cooperation, especially in 1999, 2003–2008, and 2011. The difference in the political system will directly lead to many differences in various management systems and social development among countries, which may become one of the attractions for tourists. The positive effect of the difference in the political system does not support Hypothesis 1.

Governance difference negatively impacts tourism cooperation. Specifically, it has significant negative effects on tourism cooperation from 1998 to 2011. After 2011, the significantly negative effects disappeared. With the negative effects, Hypothesis 2 is supported. As an important guarantee for the tourism industry [71,72], the governance capacity affects the performance of tourism to a certain extent. The huge governance difference between the two countries is not conducive to tourism cooperation between the two countries.


**Table 2.** Quadratic Assignment Procedure (QAP) regression analysis.

\*\*\* Statistically significant at 0.001 level; \*\* statistically significant at 0.01 level; \* statistically significant at 0.05.

Regarding the income difference factor, we observed that income difference has negative effects on tourism cooperation with a significant negative impact on tourism cooperation in most years (1998, 2000–2003, 2005, 2007, 2010, and 2012–2017). Hypothesis 3 is supported. Income is still a major part restricting the generation of tourism activities.

According to Table 2, the difference in consumption level has a complicated influence on tourism cooperation. The difference in consumption level presents a positive effect on tourism cooperation before 2006, especially having a significantly positive effect in 2001, 2002, 2003, and 2005. Additionally, it has a significant negative influence on tourism cooperation after 2008. Based on this complex phenomenon, Hypothesis 4 is not supported.

Regarding the influence of security difference, the positive effects on tourism cooperation were presented in 2012 and 2015–2017, supporting Hypothesis 5. A safe environment is a necessary condition for tourism activities. Tourists tend to choose safe destinations. As a result, security difference moderately promotes tourism cooperation.

Population density difference has a positive effect on tourism cooperation, especially in 2005 and 2017. According to the positive effect, Hypothesis 6 is supported. The difference in population density between the two countries means the difference in the tourism environment, which is also an important thrust for tourism cooperation.

Language difference has a positive effect on tourism cooperation, although these effects are not significant. Therefore, Hypothesis 7 is not supported. The positive effect reveals that language difference is no longer an obstacle for international tourism. In the international tourism environment, language difference represents the cultural difference, which becomes one of the attractions for tourists.

Based on the above analysis, the current study constructed the formation mechanism of tourism cooperation in China-ASEAN countries (Figure 3). We observed that political system difference, security difference, population density difference, and language difference jointly promote tourism cooperation. However, governance difference, income difference, and consumption level difference are obstacles for tourism cooperation presently.

**Figure 3.** Formation mechanism of China-ASEAN tourism cooperation.

#### **5. Discussion and Implications**

#### *5.1. Conclusion*

To summarize, this study measured the tourism cooperation ties of China-ASEAN countries using the modified gravity model, examined the structure of tourism cooperation through the social network analysis method, and identified the factors affecting tourism cooperation networks by employing the QAP analysis method. The results of this study contribute to tourism cooperation with several implications.

On the one hand, this study found that tourism cooperation in China-ASEAN countries has obvious structural characteristics. The scale of tourism cooperation is still small, even though the relationships in tourism cooperation for China-ASEAN are getting closer. The rapid growth of tourism in ASEAN countries can reflect the close relationship in tourism cooperation. Considering the cooperation model, the role of intermediaries has been gradually reduced since the autonomy of the network is gradually strengthening. Direct cooperation becomes a popular cooperation type that can support tourists and improve cooperation effects. As for the cooperation efficiency, the overall efficiency of the network is decreasing due to some countries not having as powerful as an influence on cooperative networks as before.

On the other hand, this study investigated the factors for tourism cooperation by using the QAP analysis. The differences in governance and income have negative effects on tourism cooperation. In detail, before 2011, governance difference has a significantly negative effect on tourism cooperation. Furthermore, the significant negative effect disappeared after 2011. Income difference has a negative effect on tourism cooperation; however, this negative effect is significant only in certain years. However, differences in the political system, security, population density, and language positively affect tourism cooperation. Besides, consumption level difference has a positive effect before 2006 and a negative effect after 2006. In detail, we analyzed and discussed these conclusions separately.

#### *5.2. Discussion*

Concerning the positive effects, political system difference promotes tourism cooperation. In line with previous studies, this study verified the influence of the political system on tourism development [67–69]. The difference in the political system can directly lead to many differences in various management systems and social development among countries, which may become one of

the attractions for tourists. For tourists, it is the essence of tourism to seek the environment different from the residential area. Therefore, political system difference becomes one of the determinants for tourists to select a destination. Additionally, increased crime rates are the problem that ASEAN countries currently face concerning tourism [19]. A safe environment has become an important basis for tourists from ASEAN countries to select their destinations. Thus, the security difference positively affects tourism cooperation. Peng et al. [112] stated that population size has a positive effect on tourism demand, which supports the positive effect of population density difference on cooperation. The higher the population density in a region, the higher the arrivals from that region [113]. Thus, tourist flows are more likely to occur. Even though language difference does not have a significant impact on tourism cooperation, we observed the potential positive effect of the language difference. Basala and Klenosky [114] noted tourists prefer to choose countries and destinations that speak the same language, which is more convenient for them. However, language difference not only represents a different official language but also represents deep cultural differences, which positively promotes international tourism activities [115–117]. Therefore, how to explore cultural differences, attract tourists, and promote tourism cooperation has become an important issue of China-ASEAN tourism cooperation.

Regarding the negative effects, governance difference negatively affects tourism cooperation. Governance is regarded as an essential factor for a country to develop tourism [118] because strengthening governance ability has become a prerequisite in terms of promoting tourism development [119]. However, governance difference may cause differences in the government system, lifestyle, and social rules, which may lead to discomfort and inconvenience for tourists from other countries. It is difficult for tourists to visit countries with many differences. Therefore, governance difference negatively affects tourism cooperation. As a supportive factor for tourism activities, the income directly determines whether tourism activities can be realized. As for the international tourism market, tourist flows are affected by economic factors such as price, income, and exchange rate [36,73–76]. It is difficult for tourists from the two countries to visit each other if the income difference of tourists in the two countries is huge. Additionally, most countries may expect to cooperate with countries with strong economies. Therefore, it is reasonable to find that income difference may inhibit tourism cooperation.

This research discussed the complex effects of consumption level difference. First, consumption level difference appears to have a positive impact on tourism cooperation, because the difference in consumption level mainly means the difference in tourism cost. Tourism activities belong to consumption activities, and the cost is an important factor in restricting tourism activities. Differences in consumption levels express differences in travel costs. Consequently, tourists may travel to countries with lower consumption levels, which contributes to tourism cooperation. With the development of tourism and the increase in tourists' income, tourists are more concerned about quality than cost. For tourists, low tourism costs may connote low tourism quality, which would reduce tourism demand [120]. Thus, consumption level difference negatively impacts tourism cooperation presently.

#### *5.3. Implications*

According to the findings of this study, this paper found that tourism cooperation in China-ASEAN countries can be strengthened from the following aspects.

On the one hand, we need to adjust the structure of cooperation and strengthen cooperation links. First, we need to expand the tourism cooperation size in responding to its small scale. At present, China-ASEAN tourism cooperation needs to be expanded. Therefore, we can promote cooperation in the tourism market, passenger flow, information, culture, and marketing to expand the scale of tourism cooperation. Second, we need to adjust the cooperation model. Direct dialogue and cooperation between governments need to be encouraged, and tourism cooperation between countries should be strengthened through the signing of memoranda and cooperation agreements between governments. Additionally, tourism enterprises of the countries also need to actively cooperate. Enterprises and governments can jointly promote tourism cooperation. Third, cooperation efficiency needs to be

improved. For the entire network, we need strong leadership to promote the development of the network. To overcome the dilemma, it is important to cultivate the network leaders while improving their network cooperation efficiency. It is encouraged to present the demonstrative and leading role to promote close cooperation between the entire network in China-ASEAN countries.

On the other hand, we need to identify the promotion factors and promote cooperation. We can improve the promoting effect of positive factors such as political system difference and security difference. First, we should reinforce the attractions of political system difference. We can regard a series of differences formed by the regime differences as tourist attractions to induce tourists by developing corresponding tourism products and routes. Second, we must create a safe environment for tourists and strengthen the security of tourists. Third, we should emphasize the role of cultural attractions by using exemplary culture as an important way to attract international tourists. Additionally, we should suppress the negative effects of negative factors. The difference in governance is an important factor hindering tourism cooperation. Therefore, countries should focus on strengthening their governance capacity to create a convenient tourism environment for tourists.

Even though this study revealed the structure of tourism cooperation in China-ASEAN countries and empirically tested the influencing factors of the network, it has certain limitations. First, this study only explored the characteristics of tourism cooperation after the basic formation of ASEAN's ten member countries. However, ASEAN was formally established in 1967. It is of great significance to explore the historical experience and evolution of tourism cooperation among ASEAN countries for adjusting the cooperation structure in the future. Second, additional important driving factors for cooperation networks exist and are worth exploring in the future. It is acknowledged that this study attempted to explore the important impact of the political system on tourism cooperation [121]. Other political influences are issues we leave for future research. It is necessary to consider the direction of tourism cooperation through the direction of tourist flow to reveal cooperation characteristics in future research.

**Author Contributions:** J.Y. analyzed the data and wrote the original draft. Y.B. designed the research model, analyzed the data, reviewed, and editing the paper. Y.J. collected and analyzed the data. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

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

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Sustainability* Editorial Office E-mail: sustainability@mdpi.com www.mdpi.com/journal/sustainability

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18