Inbound Tourism as a Driving Force for Regional Innovation: A Spatial Impact Study on China

Abstract

Cross-border flows of people, capital, and information along with inbound tourism flows can act as an important vehicle that benefits the innovation system in tourism destination areas. This study addresses the unintended but far-reaching impact of international tourism by focusing on the influence of inbound tourism on regional innovation in China. Data from 30 Chinese provinces for the years 2003–2012 are used for the empirical analysis, employing a spatial panel data model. The results show that inbound tourism may be a new and powerful driving force for regional innovation, while the effect of inbound tourism on technological innovation appears to be weaker than that on social innovation. Our findings also show that a higher market percentage of domestic tourism may weaken the impact of inbound tourism. Furthermore, the impact of inbound tourism on innovation tends to be relatively stronger in the richer and more internationally oriented provinces of China.

Keywords

inbound tourism regional innovation absorptive capability spatial panel data China

Introduction

According to the Endogenous Growth Theory, innovation - conceived of as the generation, acceptance, and implementation of new ideas, products, processes, or services (Expósito-Langa, Molina-Morales, and Capó-Vicedo 2011) - is regarded as the fundamental driving force of economic growth (Solow 1957; Grossman and Helpman 1991). In the light of the globalization and information-economic era, national and regional innovation systems tend to become more and more comprehensive and open-ended (Cooke 2004). Clearly, in the context of emerging economies, internationalization - rather than internal or domestic learning - is the more important way to upgrade innovation capabilities (Piperopoulos, Wu, and Wang 2017). Meanwhile, several studies show that knowledge spillovers are greater in regions with cultural diversity (Audretsch, Dohse, and Niebuhr 2010), and that host–guest social contacts can significantly reduce the perceived cultural distance (Fan et al. 2017). Breaking the barrier of distance, inbound tourism acts as a nexus of the destination and external resources, thus bringing many kinds of social resources to the regional innovation system of a destination area.

As a service-intensive industry composed of heterogeneous agents and activities scattered in time and space (Aldebert, Dang, and Longhi 2011), international tourism essentially engenders a large-scale exchange of ideas and information among people (Dieke 2003). The international mobility and open communication associated with inbound tourism development will benefit the social capital of the region, which is of great significance for the improvement of regional innovation capability (Yoon et al. 2015). Moreover, the distinct demand of foreign tourists will encourage more commercial creativity (Boissevain 1996), thus resulting in more entrepreneurship (Chaperon and Bramwell 2013), which in turn prompts liberal atmosphere and even more positive political support for innovation (Wilson 1997; Smith 2009). Furthermore, inbound tourism can benefit the destination through the improvement of absorptive capacity (Vita and Kyaw 2016), which is crucial for the innovation activities.

It can be inferred that inbound tourism may have a potential influence on innovation. Although a great deal of research has focused on innovation in tourism industry (see Brooker and Joppe 2014; Hjalager 2015; Romão and Nijkamp 2017), the effect of inbound tourism on innovation in host regions is seldom addressed. And it is clear that the mechanism of how inbound tourism influences innovation activities can be complex and heterogeneous under different conditions. As a consequence, the research questions in this study are (1) What is the mechanism by which inbound tourism impacts regional innovation? And will absorptive capacity play a decisive mediating role in the above relationship? (2) What are the effects from inbound tourism on different types of innovation? (3) Will this effect vary under different tourism market structures? (4) Will the performance of inbound tourism differ depending on levels of regional development and trade openness?

The novel contribution of this study may be described as follows. First, the theoretical framework and empirical study of the potential influence of inbound tourism on regional innovation will lead to more in-depth insights into not only the interaction between tourism development and innovation but also the broader systemwide impact of tourism. While this topic has seldom been discussed systematically, even studies on impacts of tourism innovation on destinations economies are very limited (Hjalager 2010). What is more, the discussion on the role of innovation types, regional conditions, and tourism market structures may give an informed interpretation of the current situation of how the regional innovation system is influenced by the development of inbound tourism in China, which can help to stimulate the co-prosperity of a tourism economy and regional innovation activities and which is of great significance for the long-term sustainable relationship between tourism and regional development. Secondly, our study is a meaningful complement to previous studies on the relationship between immigration, cultural diversity, and innovation, especially in the context of developing regions. With regard to the underlying impact mechanism, the previous studies on the impact of immigration on innovation tended to ignore the dynamic benefits from knowledge exchange, inflows of capital in an open economy, greater product variety, and consumption externalities (Ottaviano and Peri 2006), which are the main mechanisms of the way inbound tourism acts on innovation. Moreover, since Jacobs (1961) argued that there will be more innovation in more diverse cities, studies on this topic have mainly focused on an American or European context (Ozgen, Nijkamp, and Poot 2012). Nevertheless, rather than immigration, inbound tourism is a more practical and appropriate way to explore the influence of cultural diversity on innovation, especially in developing countries or regions. The present paper aims, therefore, to assess the impact of inbound tourism on regional innovation, especially in the context of Mainland China.

Analysis Framework

Influence of Inbound Tourism on Regional Innovation

The spatial network structure approach

In a regional context, innovation takes usually place in a regional innovation system (Cooke, Uranga, and Etxebarria 1997) consisting of actors, resources, and activities (Little 1987). The structure of the regional network affects the flow and quality of information (Granovetter 2005). Rather than strong ties, weak ties or structural holes are significantly important to cross-border knowledge transfer and acquisition of diverse external knowledge (Hansen 1999; Burt 2004). What is more, bridging social capital, stimulated by the ties among the diverse horizontal actors (Putnam 2000), will give birth to more positive interaction and novel information (Kallio, Harmaakorpi, and Pihkala 2010).

The development of inbound tourism cannot be separated from the process of internationalization and cross-border mobility (Lanfant 1980; C. M. Hall 2004), which provides a great opportunity for international tourism destinations to build firm links to the outside world. Co-producing efforts in multiple sectors can expand the underestimation ties, while the bridging ties can act as the information conduits and create/initiate network dynamics for the destinations (Ness, Aarstad, and Haugland 2014). The integration of local actors into international knowledge flows will give birth to an efficient regional innovation system (Fritsch and Graf 2011). Furthermore, the enhanced motivation of regional marketing can also improve the information and digitization of a tourism destination. New weak ties and structural holes help to create more opportunity for the innovation system to gain more by bridging social capital with more actors and resources for innovation activities.

Consequently, the establishment of external networks can improve the innovation of a territorial industrial system, while the performance of the firms concerned will benefit from the exploitation of business linkages (L. C. Chen 2015). Simultaneously, the expected yield of the investor will be higher in an area that has greater international influence; the perceived quality of life and career development opportunity of the employee will be higher in areas with a sounder production and recreation system. A study of the Global Power City Index shows that artists and researchers, as the important individuals and even “technological gatekeepers” (Allen 1977) of regional innovation, prefer international destination cities, such as Paris, London, and New York (Institute for Urban Strategies 2014). The participation of high-level intellectual capital will undoubtedly improve innovation in the destination area, and mobility of this group of people can also benefit the spillover of knowledge and innovation to neighboring areas.

To sum up, inbound tourism may improve the regional innovation network through attracting more institutions and resources, thus expanding the regional innovation network, and bringing in more resources for regional innovation.

The heterogeneous demand mechanics

The systemic approach to innovation argues that innovation is the result of collaboration and interaction between firms and a variety of actors, including customers, governmental institutions, public research organizations, and subcontractors, etc. (Andersson and Karlsson 2004). Besides the elements on the “supply-side,” the external requirement from the demand side can be the driving force of innovation (Schmookler 1966). Economic growth can be treated as a process of self-discovery, which is driven by the new demands of the market (Hausmann and Rodrik 2003). The diversity or heterogeneous market demand will improve the technological evolution, especially in the early technological development stages (Adner and Levinthal 2001).

Because of the development of inbound tourism, the destination faces a compound demand system that consists of both the residents’ and the tourists’ demand. Obviously, these demands are different in nature. First, it is not surprising that the local souvenirs that are treasured by the foreign tourist may be regarded as worthless for the residents. Second, demands of foreign tourists, generally orientated toward leisure and hedonism (Fodness 1994), are not necessarily shared by local residents (Canavan 2016). Third, the different cultural background of international tourists results in a different consumption preference compared to the residents, even when faced with a purchase decision relating to the same commodity (Moon, Chadee, and Tikoo 2008).

One of the profound impacts of the heterogeneous demand is the stimulation of economic diversity and entrepreneurship. Self-evidently, to satisfy the demands of foreign tourists and gain more economic benefit, suppliers may try their best to adapt to the heterogeneous demand. As the destination innovation network consists of actors from multiple industries (Zach and Hill 2017), this externality will extend to more extensive commercial innovation. More than two-thirds of the top-25 mainland Chinese cities with a well-developed tourism sector are ranked as the top-25 best commercial cities by Forbes in 2012 (China Internet Watch 2013). As a result of self-selection, entrepreneurs tend to be more risk-taking, full of energy and ability, and often younger (Kloosterman and Rath 2003), which will continually stimulate economic diversity and benefit regional innovation.

Meanwhile, except for the basic spending on accommodation, food, and transportation, the highest and most important expenditure of international tourists is on culture and recreation (Hadjikakou et al. 2013; Barnini, Cracolici, and Nijkamp 2017). Spatial consumption is of great help for creating business opportunities, especially for small- and medium-sized enterprises (Saha, Su, and Campbell 2016). Along with cultural diversity caused by the inbound tourism development, the destination will develop as an attractive pole that accumulates a variety of services and information (Ottaviano and Peri 2006; Ozgen, Nijkamp, and Poot 2012), thus reducing the transaction costs both within and across production sectors (Glaeser et al. 1992) and increasing the return of physical and human capital will be higher (Quigley 1998). These continually attract more entrepreneurship and innovation. Moreover, the wide environmental and social benefits of international tourism development will give the related innovation actors a better bargaining power when intervening policy making (M. C. Hall and Allan 2008). Therefore, creativity and innovation activities can gain wider recognition and more political encouragement in an area that benefits more from international tourism development (Canavan 2016). The favorable external factors, including market and political regulations (Mowery and Rosenberg 1979), will in turn generate more entrepreneurship.

Accordingly, the development of inbound tourism leads to a dynamic and diverse demand system, thus stimulating more entrepreneurial and commercial innovation. Given the expansion of the spatial cross-border network and the evolution of the heterogeneous demand system caused by inbound tourism development, we can formulate the following research hypothesis:

Hypothesis 1: Inbound tourism development has a significant positive influence on regional innovation.

This is one of the hypotheses to be tested in the study, besides a few more research hypotheses formulated in the subsequent subsections.

The mediating effect of absorptive capacity

The dynamic knowledge and information flow process are significant for the innovation system (Baggio, Scott, and Cooper 2010). From the perspective of knowledge and learning theory, innovation activity is a process of knowledge exchange and integration (Nooteboom 2000). As the ability to identify, assimilate, and exploit external knowledge (W. M. Cohen and Levinthal 1989), the absorptive capacity can aid knowledge transfer into the regional innovation system (Kallio, Harmaakorpi, and Pihkala 2010). By managing external knowledge flows more efficiently, the absorptive capacity can promote external collaboration (Fabrizio 2009), thus helping the innovation system to benefit more from the external knowledge and technology spillovers (W. M. Cohen and Levinthal 1989; Mancusi 2008), and stimulating innovation performance (Expósito-Langa, Molina-Morales, and Capó-Vicedo 2011).

Since the actors of innovation activities are systematically engaged in collective learning based on the institutional “milieu” (Kallio, Harmaakorpi, and Pihkala 2010), the absorptive capacity of the innovation system depends on the combination of different cognitive dimensions. Lack of cognitive proximity will reduce knowledge transfer between different networks or actors (Nooteboom 2000). A sound foundation of shared language and values, as well as a well-structured community, will all benefit knowledge exchange (Kallio, Harmaakorpi, and Pihkala 2010), while cognitive closeness can stimulate dynamic efficiency and benefit innovation (Caragliu 2015).

International tourism activities are accompanied by the exchange of different cultures and values, which stimulates the evolution and co-option of culture (E. Cohen 1988), thus making a multicultural and tolerant social environment. At the same time, as a kind of nonstandard export (Brida et al. 2016), inbound tourism development can promote the openness level of the destination region to the whole world. The mobility, connectivity, and internationalization caused by inbound tourism will help the regional innovation system to gain more external social capital, thus promoting the exploration of knowledge and the potential absorptive capacity (Kallio, Harmaakorpi, and Pihkala 2010). A diverse and complementary knowledge background will help to create a suitable level of cognitive proximity (Boschma 2005), which is necessary for an efficient innovation system, and will also benefit the regional absorptive capacity. And the adaptive culture has a positive impact on the performance of product and service innovation (Verdu-Jover, Alos-Simo, and Gomez-Gras 2017).

Accordingly, we can formulate the following two mutually related hypotheses:

Hypothesis 2: Inbound tourism development has a significant positive influence on the absorptive capacity of the destination.

Hypothesis 3: The absorptive capacity has a significant mediating effect on the relationship of inbound tourism and regional innovation.

Effect of Different Innovation Types

Innovation takes place in a variety of ways. Besides technological innovation, there is also social innovation, for example, in the field of distribution, communication, and cooperation (Gardner, Acharya, and Yach 2007). Dynamic management, flexible organization, and networking between organizations are all social innovations, which are regarded as complementary to technological innovation (Pot and Vaas 2008). Different types of innovation activities are shaped by their particular knowledge base (B. T. Asheim and Coenen 2005). Technological innovation activities are mainly based on analytical knowledge, which requires formal and professional collaboration between research organizations; social innovation is highly related to synthetic knowledge, which is always incremental innovation based on the interactive learning and novel integration of existing knowledge (B. Asheim and Gertler 2005).

Social innovation can benefit more from inbound tourism. Firstly, the mechanism of how inbound tourism influences innovation works better when it comes to social innovation. Generally, compared with social innovation, a relatively long period of professional skill and the necessary equipment is required to achieve technological innovation. What is more, the absorptive capacity in the field of technology is difficult to improve by short-term and superficial interaction, whereas the network and demand approach cannot work without large-scale professional intellectual capital investment. However, social innovation can benefit more from cultural diversity or cognitive proximity; the effects of network and demand will emerge over a much shorter term. Second, the characteristics of international tourism intensify this distinct effect. Linkage of tourism activities with technological sectors is weaker than that with culture, creativity, and other kinds of commercial industries. The inherent industrial chain makes its effect on social innovation stronger.

Therefore, we can infer the following hypothesis:

Hypothesis 4: The effect of inbound tourism on social innovation is stronger than that on technological innovation.

Effect of the Tourism Market Structure

The tourism market of the destination region consists of domestic tourism and inbound tourism, while the effects of inbound tourism can be impaired if the destination region relies heavily on the domestic tourism market. First, it is obvious that the effect of a heterogeneous demand mechanism will be weaker, if the scale of international demand is not big enough for the supplier to customize the products when most of the customers are domestic. Consequently, the cross-border mobility and stimulating multicultural environment caused by inbound tourism are not adequately transferred to be a driving force of innovation on the demand side. Without substantial knowledge transfer, inbound tourism may fail to favor the cognitive proximity, let alone to improve the absorptive capacity of destination regions.

Second, regions with a higher domestic market rate may be areas with a lower international tourist competitiveness, which are still in the primary stage of a tourism economy with relatively nonadvanced development modes, or which profit more from the locational advantage for just an elementary tourism development within an inner-provincial or inter-provincial scope. Without a broad industry convergence, a tourism economy is confined to the catering and hospitality sector, whose contribution to the innovation system will be very limited. Therefore, this limitation of market position and development mode leads to a lower externality on the local network and a lower intrinsic motivation for innovation. As a result, we can infer the next hypothesis as:

Hypothesis 5: The effect of inbound tourism on regional innovation is weaker in destinations with a higher proportion of domestic tourists.

Effect of Territorial Preconditions of a Region

Territorial preconditions and region-specific factors are significant for knowledge creation and innovation activities (Andersson and Karlsson 2004; Fritsch and Slavtchev 2010). The performance and efficiency of regional innovation depend clearly on the intensity and frequency of the interaction among the system factors (Trippl 2010). Thus, to achieve a good regional innovation performance, it is necessary to have favorable preconditions, such as infrastructure, intellectual property protection, and human resources. Moreover, the development of absorptive capacity is history path-dependent, and the gap of innovation ability between a particular region and the developed regions influence the regional absorptive capacity remarkably (W. M. Cohen and Levinthal 1989). Consequently, we may assume that the preconditions for innovation (i.e., the innovation environment), for example, the context of economic growth or the trade openness level, may have a moderating effect on the positive influence of tourism development on regional innovation.

In a well-developed area, a good innovation environment will help to realize a more effective innovation system. Adequate budget will lead to more autonomous policy and a higher possibility of innovation success (Cooke, Uranga, and Etxebarria 1997). The existing funds and management experience are sufficient, so there is a higher level of appreciation for intellectual property. While there will likely be an innovation leakage in the area that does not have a good innovation environment.

Firstly, in a less developed area, the financial budget of the government and the ability to operate are limited. The government and the companies prefer to duplicate the successful operating model rather than to create an original innovation. Owing to lack of funds and management experience, cross-border investment and brand extension, and other forms of external investment introduction become the main approach of regional development. Direct investment and service outsourcing will even spread to public services. Consequently, entrance of advanced innovation actors will cause a higher risk of innovation leakage. Part of the achievement of research and development flows out to some transnational corporations or other foreign investors. For example, by the end of 2014, among the invention patents applied for in China, only 59.2% of the intellectual property belongs to Chinese enterprises or public bodies (State Intellectual Property Office of China 2015, 2).

Second, some senior technical workers and researchers are choosing to emigrate to developed areas, which in turn aggravates the leakage effect. Unfortunately, the inbound tourism may aggravate the situation. In a less developed area, especially a destination for ethnic tourism, there is an implicit economic inequality in the process of tourism trade. Most of the time, the dominant economic position of the main tourism origin region (North American and European areas) makes foreign tourists symbols of an advanced culture, which causes the intervention of cognition and even ideology. The demonstration effect is stronger in underdeveloped areas (Brunt and Courtney 1999), and is also relatively high among young people (Murphy 2013), which results in more skilled migration and resource outflows from the less developed area. Without proper guidance, the neocolonialism trend (Lanfant, Allcock, and Bruner 1995, p. 5) in the international tourism market will be much more pronounced.

Research has shown that a country far below the world technological frontier, although reached by advanced new knowledge produced by technological leaders, will be unable to benefit from it (Mancusi 2008). Therefore, we can formulate the following (sub) hypotheses:

Hypothesis 6: The effect of inbound tourism on regional innovation varies in destinations that have different innovation preconditions.

Hypothesis 6a: The effect of inbound tourism on regional innovation is stronger in higher-developed destinations.

The level of trade openness can also be another good proxy of a precondition. The benefits of international trade for mature economies will generate extensive and long-term capital accumulation, thus leading not only to groups of innovation actors with high innovation ability but also to a more effective demand that helps to catalyze and diffuse innovation. What is more, the early internationalized areas tend to have often a more ordered and well-functioning market, a higher awareness of intellectual property protection, and a high willingness to cooperate, thus providing the regional innovation system itself with a higher productivity.

In China, the goals of inbound tourism development used to be aimed at international diplomacy in the early years (Tisdell and Wen 1991); even nowadays, in some areas inbound tourism is being treated narrowly as a means of poverty alleviation. The old business philosophy and development model is preventing the potential positive integration of the tourism industry in the local society. In view of path dependence, a particular region’s own condition may limit the positive effect of inbound tourism on regional innovation. Nevertheless, in areas with a more mature international commercial system, high-level endowment of tourism development with cultural and creative, exhibition, real estate, financial, and international trade industries will stimulate superior profit models and substantial added value. The emerging composite business forms of a tourism economy are a source of regional intensive development and social innovation. In addition, it is clear that the international business tourism plays a remarkable role in the synergies of innovation actors within and across the regions. In this case, the marginal utility of inbound tourism in the economically self-sufficient area will be lower, and we can formulate the following subhypothesis:

Hypothesis 6b: The effect of inbound tourism on regional innovation is stronger in the destinations that have a higher level of trade openness.

To sum up, the comprehensive framework of the interaction between inbound tourism and regional innovation, to be empirically tested in the remaining part of this study, can be mapped out in the following systemic archetypal scheme (see Figure 1).

Figure 1.

Interaction between inbound tourism and regional innovation.

Research Method

Measurement

The innovation system and the public administration are relatively dependent in each province, so that a single model cannot be recommended when discussing innovation activities in China (Chung 2002; K. Chen and Guan 2011). Clearly, statistical data on innovation activity at the city level are scarce in China (see Ning, Wang, and Li 2016); therefore, to ensure a broader sample area and a longer analysis period, provincial data were collected wherever available.

The data of our study come from the Chinese Patent Statistical Yearbook, the Chinese Statistical Yearbook and the China Economic & Industry Data Database, ranging from 2003 to 2012 of 30 Chinese Mainland provinces (Tibet is excluded, because some of the important indicators are not available). The innovation capacity (IC_it) is used as a proxy for regional innovation, which is measured comprehensively by the entropy method as shown in Table 1. A meta-analysis shows that the choice of proxies for tourism (income or arrivals) may influence the results of tourism statistical studies; hence, an appropriate estimation base for both these proxies of tourism is encouraged (Castro-Nuño, Molina-Toucedo, and Pablo-Romero 2013). Therefore, the entropy method was used in the measurement of inbound tourism (T_it). Absorptive capacity is also measured by the entropy method, according to the existing studies. Two types of innovation (Tin_it and Sin_it) are measured by the related innovation output as listed in Table 1. To keep the key variables in a similar order of magnitudes, the results of the entropy calculation were expanded 10,000 times, and all variables are in logarithms.

Table 1.

Measurement of the Key Variables.

Concepts	Measurement	Variables	Mean	SD	Median	Upper Quantiles
Inbound Tourism	Entropy Method based on: Foreign tourism reception Foreigner tourism income^a	lnT_it	3.136	0.835	3.043	2.509
Innovation Capacity	Entropy Method based on: Innovation Input: New fix assert investment of R&D R&D intensity (proportion of R&D investment in real Gross Domestic Product) R&D expenditure Employee of R&D Innovation Output: Granted patent per person^b Application for patent Technological market turnover^c Innovation Environment: real GDP per capita Trade Openness level (proportion of total imports and exports in the real GDP) FDI intensity (proportion of investment from foreign-invested enterprises in real GDP) Human resources (proportion of people with high school education level in the population above 6) Graduate students of university or college	lnIC_it	3.448	0.324	3.372	3.205
Absorptive Capacity	Entropy and Perpetual Inventory Method based on^d: Absorpt1= pg(t – 1)( 1 – δ)+Δpg(t) Absorpt2= hr(t – 1)( 1 – δ)+Δhr(t)	lnAC_it	3.419	0.404	3.369	3.113
Technological Innovation	Patent application of invention per 100 000 persons	lnTin_it	4.258	1.270	4.065	3.269
Social Innovation	Total patent application of utility model and design per 100 000 persons	lnSin_it	5.219	1.212	5.010	4.227

Note:

Because there are seldom particular official statistics for income from foreign tourists at the province level (according to the China Economic & Industry Data Database), the data on foreign tourism income is calculated based on the inbound tourism income, number of foreign tourists, and number of inbound tourists of each province: foreign tourism income = inbound tourism income × (number of foreign tourists /number of inbound tourists).

According to the provisions of the State Intellectual Property Office of the People’s Republic of China (http://www.sipo.gov.cn/zlsqzn/), the examination and review of patent applications is a long process. With regard to an investment patent, it takes 3 years for an application to be granted; with regard to the utility model and design patent, the whole process can be completed within one year. On average, we assume that the innovation output of a particular year is reflected in the data on patents granted in the next year. So we use the data on granted patents in the year t+1 to measure the Granted Patent in the year t.

All economic indicators that are influenced by prices are deflated based on the Consumer Price Index of 2003.

Given the research approach of Borensztein, De Gregorio, and Lee (1998) and Caragliu (2015), our study measures Absorptive Capacity on the basis of the accumulation of human capital and granted patents, which refers to both the potential absorptive capacity and the realized absorptive capacity as distinguished by Zahra and George (2002). The accumulation of human capital (hr_t) and granted patents (pg_t) are calculated as follows:

accumulation of human capital: hr_t = hr_t–1*(1 – δ)+Δhr_t

accumulation of granted patents: gp_t =gp_t–1*(1 – δ)+Δgp_t,

where: δ is the average discount rate, and δ =5%; hr is the higher education graduate; Δhr_t is admissions to higher education in year t; gp is the granted patent, Δpg_t is the increase in the number of patents granted in year t.

To test the moderating effects of the tourism market structure and the preconditions of regions, the various provinces in China were subdivided into two groups according to the relative level of domestic tourism, economic development, and trade openness levels. The tourism market structure was proxied by the proportion of domestic tourism income in total tourism income. Inbound tourism in provinces with a higher (lnTsh_it) and lower (lnTsl_it) domestic tourism proportion was obtained according to the relative level of domestic tourism proportion (str_it) compared to the average level (str₀) as

\begin{array}{l} l n T s h_{i t} = {\begin{matrix} l n T_{i t}, & s t r_{i t} \geq s t r_{0} \\ 0, & s t r_{i t} < s t r_{0} \end{matrix} \\ l n T s l_{i t} = {\begin{matrix} 0, & s t r_{i t} \geq s t r_{0} \\ l n T_{i t}, & s t r_{i t} < s t r_{0} \end{matrix} \end{array}

The absorptive capacity was also subdivided into lnACsh_it and lnACsl_it,. These expressions refer to the Absorptive capacity in provinces with a higher or lower domestic tourism proportion, respectively.

\begin{array}{l} l n A C s h_{i t} = {\begin{matrix} l n A C_{i t}, & s t r_{i t} \geq s t r_{0} \\ 0, & s t r_{i t} < s t r_{0} \end{matrix} \\ l n A C s l_{i t} = {\begin{matrix} 0, & s t r_{i t} \geq s t r_{0} \\ l n A C_{i t}, & s t r_{i t} < s t r_{0} \end{matrix} \end{array}

By the same method, we obtain lnTeh_it and lnTel_it (inbound tourism in provinces with a stronger or weaker economic base), lnToh_it and lnTol_it (inbound tourism in provinces with a higher or lower trade openness level), as well as lnACeh_it and lnACel_it (the absorptive capacity in provinces with a stronger or weaker economic base), and lnACoh_it and lnACol_it (the absorptive capacity in provinces with a higher or lower trade openness level).

Pretests and Models

Levin-Lin-Chu and Breitung tests were conducted to test the stability of the variables; the results rejected the existence of a unit root. (P values of lnT_it, lnIC_it, lnAC_it, lnTin_it, lnSin_it of these two tests are, respectively, 0.0000, 0.0000, 0.0376, 0.0000, 0.0000, and 0.019, 0.0121, 0.0141, 0.0035, 0.0000.) Consistent with the existing studies (Wu 2006; Tian, Wang, and Chen 2010; Yang and Wong 2012), the presence of spatial autocorrelation, measured by the Moran’s I statistic, shows that a significant spatial autocorrelation exists among the key variables (see Table 2). And the results are stable even with different kinds of a weight matrix and a different distance cut-off point.

Table 2.

Spatial Autocorrelation Tests of the Key Variables.

Year	lnT	lnIC	lnAC	lnTin	lnSin
2003	0.061***	0.097***	0.064***	0.116***	0.118***
2004	0.054***	0.111***	0.067***	0.125***	0.124***
2005	0.059***	0.123***	0.069***	0.104***	0.113***
2006	0.066***	0.130***	0.071***	0.109***	0.113***
2007	0.073***	0.125***	0.074***	0.117***	0.110***
2008	0.104***	0.130***	0.072***	0.114***	0.119***
2009	0.106***	0.132***	0.076***	0.104***	0.121***
2010	0.104***	0.138***	0.079***	0.100***	0.133***
2011	0.098***	0.133***	0.083***	0.098***	0.130***
2012	0.097***	0.141***	0.088***	0.092***	0.125***

Note: Values are the Moran’s I of variables in related years calculated based on the inverse distance spatial weight matrix.

***

Significance at the 5% level.

Therefore, spatial panel data models were employed when testing the hypotheses of our study. The Spatial Durbin Model (SDM) is adopted as the initial model of the estimation. And the results of the Lagrange multiplier and likelihood ratio tests were considered (Elhorst 2014; Z. Chen and Haynes 2015) when choosing the final form for each model from the Spatial Error Model (SEM), the Spatial Lag Model (SAR), and the SDM. According to Baron and Kenny (1986), the statistical models used to test hypotheses 1, 2, and 3 are

\begin{matrix} l n {IC}_{it} = ρ_{1} \sum {w_{ij} {lnIC}_{jt}}^{} + η_{1} {lnT}_{it} + η_{2} \sum {w_{ij} {lnT}_{it}}^{} \\ + \sum {δ_{j 1} {lnX}_{it}}^{} + ζ_{1} \sum {w_{ij} {lnX}_{it}}^{} + ξ_{it}, \end{matrix}

\begin{matrix} {lnAC}_{it} = ρ_{2} \sum w_{ij} {lnAC}_{jt} + β_{1} T_{it} + β_{2} \sum w_{ij} {lnT}_{it} \\ + \sum δ_{j 2} {lnX}_{it} + ζ_{2} \sum w_{ij} {lnX}_{it} + γ_{it}, \end{matrix}

and

\begin{array}{l} {lnIC}_{it} = ρ_{3} \sum^{} w_{ij} {lnIC}_{jt} + ζ_{1} {lnT}_{it} + ζ_{2} \sum^{} w_{ij} {lnT}_{it} \\ + ζ_{3} {lnAC}_{it} + ζ_{4} \sum^{} w_{ij} {lnAC}_{jt} + \sum^{} δ_{j 3} {lnX}_{it} \\ + ζ_{3} \sum^{} w_{ij} {lnX}_{it} + θ_{it} . \end{array}

where X_it refers to the matrix of control variables. Hypothesis 4 was tested by equation (1) with lnIC_it replaced by lnTin_it or lnSin_it. Hypothesis 5 is tested by

\begin{matrix} {lnIC}_{it} = δ_{1} \sum^{} w_{ij} {lnIC}_{jt} + α_{11} {lnTsh}_{it} + α_{12} \sum^{} w_{ij} {lnTsh}_{it} \\ + α_{13} {lnTsl}_{it} + α_{14} \sum^{} w_{ij} {lnTsl}_{it} + \sum^{} δ_{j 14} {lnX}_{it} \\ + ζ_{14} \sum^{} w_{ij} {lnX}_{it} + ε_{i t}, \end{matrix}

\begin{matrix} {lnAC}_{it} = δ_{2} \sum^{} w_{ij} {lnIC}_{jt} + α_{21} {lnTsh}_{it} + α_{22} \sum^{} w_{ij} {lnTsh}_{it} \\ + α_{23} {lnTsl}_{it} + α_{24} \sum^{} w_{ij} {lnTsl}_{it} + \sum^{} δ_{j 24} {lnX}_{it} \\ + ζ_{24} \sum^{} w_{ij} {lnX}_{it} + ε_{i t}, \end{matrix}

and

\begin{matrix} {lnIC}_{it} = δ_{3} \sum^{} w_{ij} {lnIC}_{jt} + α_{31} {lnTsh}_{it} + α_{32} \sum^{} w_{ij} {lnTsh}_{it} \\ + α_{33} {lnTsl}_{it} + α_{34} \sum^{} w_{ij} {lnTsl}_{it} + α_{35} {lnAC}_{it} \\ \begin{array}{l} + α_{36} \sum^{} w_{ij} {lnAC}_{jt} + \sum^{} δ_{j 34} {lnX}_{it} \\ + ζ_{34} \sum^{} w_{ij} {lnX}_{it} + ε_{i t}, \end{array} \end{matrix}

Similar models are used when we test hypothesis 6.

Given the fact that the inbound tourism level is sensitive to many economic and social characteristics, in order to achieve a more reasonable estimation of the models and to avoid the negative effect of multicollinearity, only R&D Intensity (lnINTEN_it), FDI Intensity (lnFDI_it), and Population (above 6 years old) of Uneducated (lnUEDU_it) were controlled when the dependent variable is lnIC_it. lnTin_it or lnSin_it. Employees in R&D (per 1000 persons) (lnEMP_it), Granted patent per 10,000 persons (lnPG_it), and Technological market turnover (lnTMT_it), as well as lnUEDU_it, and lnINTEN_it were controlled for when the dependent variable is lnAC_it.

Empirical Research Findings

Hypotheses Tests

Econometric modeling methods were employed to test our hypotheses. To avoid the potential endogeneity and heterogeneity problem, fixed effect models (Baltagi 2001) and generalized method of moments estimation based on Arellano and Bond (1991) were used for the spatial panel regressions. Models 1 and 2 in Table 3 show that both the absorptive capacity and the innovation capacity are significantly impacted by inbound tourism. Based on model 3, the influence of inbound tourism appears to be weaker but still significant, after controlling for the contribution of absorptive capacity to regional innovation capacity. Therefore, the impact of inbound tourism on regional innovation capacity is partly mediated by the absorptive capacity approach. According to Biesanz, Falk, and Savalei (2010), the indirect and direct effect of inbound tourism on the innovation capacity is 0.045 (0.114*0.396) and 0.077. Consequently, hypotheses 1, 2, and 3 are all confirmed. This conclusion supports not only the absorptive capability approach but also the network structure and heterogeneous demand approaches. The interaction of diversity of culture caused by inbound tourism may reduce the cognitive obstacle and promote the ability to identify, assimilate, and exploit external knowledge, which, in turn, improves the regional innovation activities.

Table 3.

Influence of Inbound Tourism and the Market Structure Effect.

	IC_it	AC_it	IC_it	Tin_it	Sin_it	IC_it	AC_it	IC_it
	Hypothesis 1	Hypothesis 2	Hypothesis 3	Hypothesis 4		Hypothesis 5
Dependent Variables	Model 1SDM	Model 2 SAR	Model 3SDM	Model 4 SAR	Model 5 SEM	Model 6 SAR	Model 7 SAR	Model 8 SAR
lnT_it	0.488^***	0.144^***	0.077^***	0.304^***	0.760^***
lnTsh_it						0.131^*	0.272^***	0.016
lnTsl_it						0.411^***	0.434^***	0.060^**
lnAC_it			0.396^***
lnACsh_it								0.343^***
lnACsl_it								0.396^***
lnFDI_it	0.112^*		0.127^***	0.099^**	0.032	0.113^***		0.088^***
lnUEDU_it	0.388^***	0.007	0.026	−0.236^**	−0.199^***	0.310^**	0.322^***	0.020
lnINTEN _it	0.328^***	0.453^***	0.104^*	0.303^***	0.769^***	0.192^**	−0.067	0.026
lnEMP_it		−0.237^***					0.118^***
lnPG_it		0.054					0.193^***
lnTMT_it		−0.009					−0.012
w*lnIC_it	0.491^***		0.801^***			0.461^***		0.465^***
w*lnAC_it		0.522^***					0.366^***
w*lnTin_it				0.769^***
R ²	0.880	0.940	0.969	0.965	0.673	0.918	0.950	0.972
Log.L	288.34	265.39	510.70	46.40	78.45	337.55	308.76	541.44
LM error	0.448	0.017	0.312	0.521	0.056	0.214	0.841	0.106
LM lag	0.000	0.000	0.000	0.000	0.208	0.000	0.000	0.000
LM error robust	0.000	0.000	0.000	0.000	0.010	0.055	0.030	0.000
LM lag robust	0.000	0.000	0.000	0.000	0.046	0.000	0.000	0.000
LR	0.000	–	0.000	–	–	–	–	–
t-test	–	–	–	–	–	0.006	0.006	0.006

Notes: SDM = Spatial Durbin Model; SAR = Spatial Lag Mode; SEM = Spatial Error Model; Log.L = Log-likelihood; LM = Lagrange multiplier; LR = Likelihood ratio. Values in the rows of LM error, LM lag, LM error robust, LM lag robust, LR, and t-test are the p values of the tests. Asterisks refer, respectively, to significance at the ***1%, **5% and *10% level. To avoid a cumbersome table, the spatial effect of the control variables are not listed here. SDMs were used as the LR test is significant and Log.L can be lower than SAR or SEM, although LM tests may also be significant.

According to models 4 and 5, the impact of inbound tourism on technological innovation is weaker than that on social innovation, and therefore hypothesis 4 is supported. The knowledge transfer related to technological innovation is more professional in nature, part of which is even achieved in an enclave context. The dependence of technological innovation activity on the social environment is weaker than that of social innovation activities. However, tourism can be embedded much better in various niches of social activity; the network, demand, and absorptive capacity channels may work better when it comes to social innovation.

Models 6 to 8 show that the impact of inbound tourism on innovation capacity is weaker in provinces where domestic tourism income proportion is higher (0.131<0.411), and so does the effect on absorptive capacity (0.272<0.434). When we take the mediating effect of absorptive capacity into consideration, the influence of inbound tourism on innovation capacity is still higher in provinces where the income proportion of domestic tourism is lower, and this conclusion remains stable if we consider either the indirect or the direct effect of inbound tourism (see models 7 and 8). The direct influence of inbound tourism on innovation can even be insignificant when we control the effect of absorptive capacity (see 0.016 in model 8). The results of t-tests reflect the fact that there is significant difference between the effects of inbound tourism in different subregions. That is to say, the positive impacts on spatial network, heterogeneous demand, and absorptive capacity will be weaker when more tourism income is created by the domestic market, and hence hypothesis 5 is supported.

Table 4 shows that both the economic basis and the trade openness level may play a pivotal role in the relationship between inbound tourism and innovation. A higher economic achievement level and a higher degree of trade openness may offer better conditions and a higher possibility, not only for the successfulness of the innovation activity but also for the realization of its market value. Conversely, in a poor and self-sufficient area in China (or other developing countries), the situation of innovation leakage may be relatively severe. Models 1 and 4 show that the positive effect of inbound tourism on regional innovation is weaker and less significant in relatively poor or self-sufficient areas. Provided there is a mediating effect of absorptive capacity, inbound tourism in provinces with a lower economic level cannot benefit regional innovation directly (see model 3). That also holds in an area where the trade openness level is relatively low (see regions in model 6). It is clear that the direct effect and indirect effect of inbound tourism on regional innovation are all stronger in areas with higher economic and trade openness levels. According to results of the t-test, hypotheses 6a and 6b are all supported. In order to look further into the effect of inbound tourism on innovation in different subareas, the impacts of inbound tourism in east and west China² were also mutually compared (see models 7–9 in Table 4). Consistent with models 1–6, the regional innovation in west China, where the solidity of the economic basis and trade openness are lower, cannot benefit from inbound tourism.

Table 4.

Influence of Inbound Tourism in Different Regions.

	IC_it	AC_it	IC_it	IC_it	AC_it	IC_it	IC_it	AC_it	IC_it
Dependent Variables	Model 1SAR	Model 2 SDM	Model 3SDM	Model 4SAR	Model 5 SEM	Model 6 SDM	Model 7 SEM	Model 8 SEM	Model 9 SDM
lnTeh_it	0.103^***	0.561^***	0.068^**
lnTel_it	0.088^***	0.019	0.032
lnToh_it				0.098^***	0.613^***	0.148^***
lnTol_it				0.092^***	−0.058.	0.009
lnTe_it							0.372^***	0.776^***	0.108^***
lnTw_it							−0.159^**	0.213	−0.066^*
lnACeh_it			0.366***
lnACel_it			0.352***
lnACoh_it						0.354^***
lnACol_it						0.456^***
lnACe_it									0.208^***
lnACw_it									0.693^***
lnFDI_it	0.106^***		0.108^***	0.105^***		0.107^***	0.099^***		0.065^***
lnUEDU_it	0.019	0.041	0.106^***	0.021	0.083	0.076^**	−0.245^***	−0.309^**	0.115^***
lnINTEN_it	0.147^***	0.173	0.107^***	0.147^***	−0.000	0.140^***	0.068^*	0.081	0.057^**
lnEMP_it		−0.145			0.012			0.120
lnPG_it		0.241^*			0.191^***			0.046
lnTMT_it					0.024			−0.027
w*lnIC_it	0.789^***		0.694^***	0.795^***		0.698^***			0.393^***
w*lnAC_it		0.360^***
R²	0.965	0.935	0.971	0.965	0.938	0.970	0.923	0.898	0.969
Log.L	450.69	274.62	527.20	446.58	220.23	504.26	341.36	150.15	511.63
LM error	0.040	0.919	0.985	0.046	0.013	0.303	0.000	0.000	0.135
LM lag	0.000	0.044	0.000	0.000	0.762	0.000	0.017	0.116	0.003
LM error robust	0.000	0.000	0.000	0.000	0.009	0.000	0.000	0.000	0.977
LM lag robust	0.000	0.000	0.000	0.000	0.411	0.000	0.000	0.004	0.010
LR	–	0.000	0.000	–	–	0.000	–	–	0.000
T-test	0.006	0.006	0.006	0.030	0.030	0.003	0.006	0.000	0.000

Notes: lnTe_it, lnTw_it, lnACe_it, and lnACw_it refer to, respectively, inbound tourism in east provinces, inbound tourism in west provinces, absorptive capacity in east provinces, and absorptive capacity in west provinces. SDM = Spatial Durbin Model; SAR = Spatial Lag Mode; SEM = Spatial Error Model; Log.L =Log-likelihood; LM = Lagrange multiplier; LR = Likelihood ratio. Values in the rows of LM error, LM lag, LM error robust, LM lag robust, LR, and t-test are the p values of the tests. Asterisks refer, respectively, to significance at the ***1%, **5% and *10% level. To avoid a cumbersome table, the spatial effect of the control variables are not listed here. SDM were used as the LR test is significant, and Log.L can be lower than SAR or SEM, although LM tests may also be significant.

Sensitivity Tests

In light of the above results, we conclude that all hypotheses formulated above appear to be supported. To test the robustness of this conclusion, a number of different methods are used. First, a new weight matrix based on the inverse squared distance was used to test the analysis framework, and all the hypotheses appear to be supported. Secondly, our sample was divided into two parts depending on the relative level of real GDP per capita or trade openness, so as to revisit hypothesis 6; the (sub)hypotheses appear to be also all supported. Finally, an alternative way to measure the inbound tourism was used as well. The calculation of foreign tourism income based on the number of foreign tourists, average days of stay of foreign tourists, and average expenditure per day of foreign tourists was used to replace the data of foreign tourism income in the empirical analysis above. Even though all the hypotheses are supported, the results also show that in relatively poor and more self-sufficient areas, inbound tourism may not benefit regional innovation significantly,³ which is consistent with the double-checked analyses above. Therefore, the robust and critical conclusion can be drawn that inbound tourism may play a positive role in regional innovation, and that this impact may be less valid under some specific regional conditions.

Conclusion and Discussion

This study has designed a framework to analyze the effect of inbound tourism on regional innovation by applying various approaches based on network structure, heterogeneous demand, and absorptive capacity improvement. The mediating effect of absorptive capacity, as well as the moderating effect that innovation types, tourism market structure, and preconditions of destination areas exert, were also considered. Our empirical analysis has clearly supported the prior specified theoretical hypotheses. The main conclusions of this study are as follows: first, inbound tourism has both a direct and indirect impact on regional innovation, while absorptive capacity has a significant mediating effect in this relationship; second, the positive effect from inbound tourism on social innovation is stronger than that on technological innovation; third, the effect of inbound tourism may be weaker in an area where domestic tourism is more dominant; and finally, both the economic and trade openness level may form the bottlenecks for the validation of the impact of inbound tourism on regional innovation; thus, the impact of inbound tourism in west China can even be negative. Consequently, inbound tourism can benefit regional innovation, while a rise in economic performance and trade openness can improve this positive impact.

Recently, it has been argued that the characteristics of the tourism industry benefit regional innovation in various respects. First, to promote the tourism destination and to improve the tourists’ experience value, tourism enterprises and authorities have spared no effort to push and practice geographic informatization, mobile electronic commerce, and many other kinds of social network innovation directly. As a result, this decline in information asymmetry strengthens the linkage between the region concerned and the outside world, thus helping to expand the regional innovation network. Second, as a kind of compounded subjective experience, favorable tourism activities strongly depend on co-production and customization, thus making the tourism industries very sensitive to people’s continuously changing needs. For example, whereas previously, American Express promoted the adaptation of credit cards around the world by its tourism business (M. C. Hall and Allan 2008), more recently, Airbnb and Uber and other tourism companies have led the business model innovation in the sharing economy era. Given the wide industrial relevance of the tourism industry, the leading effect of the tourism industry in commercial innovation may be an increasingly positive externality to regional innovation. Third, as a series of temporary consumption activities outside the daily living area, the boom in the tourism industry and the universal use of mobile consumption applications have benefited each other in recent years. Studies also show that experience of tourism social media usage can influence tourists’ social interacting behaviors (Mkono and Tribe 2016). Newly emerging data on demand and behavioral preferences may create a better foundation for commercial practitioners to assess consumer demands, especially when combined with Big Data technology and related statistical methods.

Clearly, the present research has some limitations; future studies may be conducted to cope with these limitations. First, restricted by the availability of data, we were not able to distinguish business tourists and leisure tourists at the province level when preforming a panel analysis in the context of China. Time series analysis of the effect of inbound tourism on the national innovation system may be carried out based on the statistics of inbound tourists with different motivations. By considering the different effects and mechanisms of international business travel and leisure activities, our theoretical framework can certainly be extended. Second, although the economic basis and trade openness level are also potential impact factors of regional innovation, they are not controlled directly in our empirical models because of the strong correlation between these factors and inbound tourism. Therefore, further empirical case study research may focus on particular regions, which may be a good alternative to perform a more thorough analysis of the relationships between inbound tourism and regional innovation. Third, given the comparability of regions, this study only considers the provinces of mainland China. More extensive analyses of regions in other developing or emerging countries are needed, so as to test the reliability robustness and generality of the conclusions above.

Footnotes

Acknowledgements

We thank the editors and reviewers for the critical comments. We also thank Patricia Ellman for her linguistic and editorial assistance.

Author’s Note

Peter Nijkamp is also affiliated to JADS (Jheronimus Academy of Data Science), Hertogenbosch, The Netherlands.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: We acknowledge financial support from National Natural Science Foundation of China (No. 41801138, No. 71773101), Humanities and Social Sciences of Ministry of Education Planning Fund (NO.16YJA790018), as well as Fundamental Research Funds for the Central Universities, China University of Geosciences (Wuhan) (No. CUG180619 and “Energy Environment and Social Management”).

Notes

ORCID iD

Jingjing Liu

Author Biographies

Jingjing Liu works in China University of Geosciences, Wuhan, China. Her research fields are tourism economics and consumer behaviour, especially focuses on innovation, livability and other comprehensive influence of tourism development on the destination, as well as tourists behaviour related to authenticity epericence and mental accounting.

Peter Nijkamp, Dr. in regional economics, Erasmus University, Rotterdam, is working in Tinbergen Institute and JADS (Jheronimus Academy of Data Science), the Netherlands, and Department of Geography, Adam Mickiewicz University, Poland. In the past years, he has focused his research in particular on new quantitative methods for policy analysis, as well as on spatial-behavioral analysis of economic agents.

References

Adner

Levinthal

. 2001. “Demand Heterogeneity and Technology Evolution: Implications for Product and Process Innovation.” Management Science 47 (5): 611–28.

Aldebert

Dang

R. J.

Longhi

. 2011. “Innovation in the Tourism Industry: The Case of Tourism@.” Tourism Management 32 (5): 1204–13.

Allen

. 1977. Managing the Flow of Technology: Technology Transfer and the Dissemination of Technological Information within the R&D Organization. Cambridge, MA: MIT Press.

Andersson

Karlsson

. 2004. “The Role of Accessibility for the Performance of Regional Innovation Systems.” In Knowledge Spillovers and Knowledge Management, edited by Karlsson

Flensburg

Hörte

Sven-Åke

, 283–309. Cheltenham, UK: Edward Elgar.

Arellano

Bond

. 1991. “Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations.” Review of Economic Studies 58:277–97.

Asheim

Gertler

M. S

. 2005. “The Geography of Innovation: Regional Innovation Systems.” In The Oxford Handbook of Innovation, edited by Fagerberg

Mowery

Nelson

, 291–317. Oxford: Oxford University Press.

Asheim

B. T.

Coenen

. 2005. “Knowledge Bases and Regional Innovation Systems: Comparing Nordic Clusters.” Research Policy 34 (8): 1173–90.

Audretsch

D. B.

Dohse

Niebuhr

. 2010. “Cultural Diversity and Entrepreneurship: A Regional Analysis for Germany.” Annals of Regional Science 45:55–85.

Baggio

Scott

Cooper

. 2010. “Network Science: A Review Focused on Tourism.” Annals of Tourism Research 37 (3): 802–27.

10.

Baltagi

B. H.

2001. Econometric Analysis of Panel Data. Chichester, UK: Wiley.

11.

Barnini

Cracolici

M. F.

Nijkamp

. 2017. “Place-Based Attributes and Spatial Expenditure Behavior in Tourism.” Journal of Regional Science 57 (2): 218–44.

12.

Baron

R. M.

Kenny

D. A

. 1986. “The Moderator–Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic, and Statistical Considerations.” Journal of Personality and Social Psychology 51 (6): 1173.

13.

Biesanz

J. C.

Falk

C. F.

Savalei

. 2010. “Assessing Mediational Models: Testing and Interval Estimation for Indirect Effects.” Multivariate Behavioral Research 45 (4): 661–701.

14.

Boissevain

1996. Coping with Tourists: European Reactions to Mass Tourism. Oxford: Berghahn Books.

15.

Borensztein

De Gregorio

Lee

J. W

. 1998. “How Does Foreign Direct Investment Affect Economic Growth?” Journal of International Economics 45 (1): 115–35.

16.

Boschma

2005. “Proximity and Innovation: A Critical Assessment.” Regional Studies 39 (1): 61–74.

17.

Brida

J. G.

Cortes-Jimenez

Pulina

. 2016. “Has the Tourism-Led Growth Hypothesis Been Validated? A Literature Review.” Current Issues in Tourism 19 (5): 394–430.

18.

Brooker

Joppe

. 2014. “Developing a Tourism Innovation Typology: Leveraging Liminal Insights.” Journal of Travel Research 53 (4): 500–8.

19.

Brunt

Courtney

. 1999. “Host Perceptions of Sociocultural Impacts.” Annals of Tourism Research 26 (3): 493–515.

20.

Burt

R. S.

2004. “Structural Holes and Good Ideas.” American Journal of Sociology 110 (2): 349–99.

21.

Canavan

2016. “Tourism Culture: Nexus, Characteristics, Context and Sustainability.” Tourism Management 53:229–43.

22.

Caragliu

2015. “The Economics of Proximity: Regional Growth, beyond Geographic Proximity.” PhD diss., Vrije Universiteit Amsterdam, Amsterdam.

23.

Castro-Nuño

Molina-Toucedo

J. A.

Pablo-Romero

M. P

. 2013. “Tourism and GDP: A Meta-analysis of Panel Data Studies.” Journal of Travel Research 52 (6): 745–58.

24.

Chaperon

Bramwell

. 2013. “Dependency and Agency in Peripheral Tourism Development.” Annals of Tourism Research 40:132–54.

25.

Chen

Guan

. 2011. “Mapping the Functionality of China’s Regional Innovation Systems: A Structural Approach.” China Economic Review 22 (1): 11–27.

26.

Chen

L. C.

2015. “Building Extra-regional Networks for Regional Innovation Systems: Taiwan’s Machine Tool Industry in China.” Technological Forecasting and Social Change 100:107–17.

27.

Chen

Haynes

K. E

. 2015. “Multilevel Assessment of Public Transportation Infrastructure: A Spatial Econometric Computable General Equilibrium Approach.” Annals of Regional Science 54 (3): 663–85.

28.

China Internet Watch. 2013. “Top Cities on Mainland China by Forbes.” http://www.chinainternetwatch.com/1895/top-cities-on-mainland-china-by-forbes/ (accessed August 28, 2016).

29.

Cohen

1988. “Authenticity and Commoditization in Tourism.” Annals of Tourism Research 15 (3): 371–86.

30.

Cohen

W. M.

Levinthal

D. A

. 1989. “Innovation and Learning: The Two Faces of R&D.” Economic Journal 99:569–96.

31.

Cooke

2004. “Evolution of Regional Innovation Systems: Emergence, Theory, Challenge for Action.” In Regional Innovation Systems 2, edited by Cooke

Heidenreich

Braczyk

H.J.

, 1–18. London: Routledge.

32.

Cooke

Uranga

M. G.

Etxebarria

. 1997. “Regional Innovation Systems: Institutional and Organisational Dimensions.” Research Policy 26 (4): 475–91.

33.

Chung

2002. “Building a National Innovation System through Regional Innovation Systems.” Technovation 22:485–91.

34.

Dieke

P. U. C.

2003. “Tourism in Africa’s Economic Development: Policy Implications.” Management Decision 41 (3): 287–98.

35.

Elhorst

J. P.

2014. “Spatial Panel Models.” In Handbook of Regional Science, edited by Fischer

M. M.

Nijkamp

, 1637–52. Berlin: Springer.

36.

Expósito-Langa

Molina-Morales

F. X.

Capó-Vicedo

. 2011. “New Product Development and Absorptive Capacity in Industrial Districts: A Multidimensional Approach.” Regional Studies 45 (3): 319–31.

37.

Fabrizio

K. R.

2009. “Absorptive Capacity and the Search for Innovation.” Research Policy 38 (2): 255–67.

38.

Fan

D. X. F.

Zhang

H. Q.

Jenkins

C. L.

Lin

P. M. C

. 2017. “Does Tourist-Host Social Contact Reduce Perceived Cultural Distance?” Journal of Travel Research 56 (8): 998–1010.

39.

Fodness

1994. “Measuring Tourist Motivation.” Annals of Tourism Research 21 (3): 555–81.

40.

Fritsch

Graf

. 2011. “How Sub-national Conditions Affect Regional Innovation Systems: The Case of the Two Germanys.” Papers in Regional Science 90 (2): 331–53.

41.

Fritsch

Slavtchev

. 2010. “How Does Industry Specialization Affect the Efficiency of Regional Innovation Systems?” Annals of Regional Science 45 (1): 87–108.

42.

Gardner

C. A.

Acharya

Yach

. 2007. “Technological and Social Innovation: A Unifying New Paradigm for Global Health.” Health Affairs 26 (4): 1052–61.

43.

Glaeser

E. L.

Kallal

H. D.

Scheinkman

J. A.

Shleifer

. 1992. “Growth in Cities.” Journal of Political Economy 100 (6): 1126–52.

44.

Granovetter

2005. “The Impact of Social Structure on Economic Outcomes.” Journal of Economic Perspectives 19 (1): 33–50.

45.

Grossman

G. M.

Helpman

. 1991. “Quality Ladders in the Theory of Growth.” Review of Economic Studies 58 (1): 43–61.

46.

Hadjikakou

Chenoweth

Miller

Druckman

Gang

. 2013. “Rethinking the Economic Contribution of Tourism: Case Study from a Mediterranean Island.” Journal of Travel Research 53 (5): 610–24.

47.

Hall

C. M.

2004. “Tourism and Mobility.” Paper Presented at the Council for Australasian Tourism and Hospitality Education (CAUTHE) Annual Conference, Brisbane.

48.

Hall

M. C.

Allan

. 2008. Tourism and Innovation. New York: Routledge.

49.

Hansen

1999. “The Search-Transfer Problem: The Role of Weak Ties in Sharing Knowledge across Organizational Subunits.” Administrative Science Quarterly 44:83–111.

50.

Hausmann

Rodrik

. 2003. “Economic Development as Self-Discovery.” Journal of Development Economics 72 (2): 603–33.

51.

Hjalager

A. M.

2010. “A Review of Innovation Research in Tourism.” Tourism Management 31 (1): 1–12.

52.

Hjalager

A. M.

2015. “100 Innovations That Transformed Tourism.” Journal of Travel Research 54 (1): 3–21.

53.

Institute for Urban Strategies. 2014. “Global Power City Index Yearbook 2014.” Tokyo.

54.

Jacobs

1961. The Death and Life of the Great American Cities. New York: Random House.

55.

Kallio

Harmaakorpi

Pihkala

. 2010. “Absorptive Capacity and Social Capital in Regional Innovation Systems: The Case of the Lahti Region in Finland.” Urban Studies 47 (2): 303–19.

56.

Kloosterman

Rath

. 2003. Immigrant Entrepreneurs: Venturing Abroad in the Age of Globalization. Berg: University of New York Press.

57.

Lanfant

M. F.

1980. “Introduction: Tourism in the Process of Internationalization.” International Social Science Journal 32 (1): 14–43.

58.

Lanfant

M. F.

Allcock

J. B.

Bruner

E. M

. 1995. International Tourism: Identity and Change, Vol. 48. London: Sage.

59.

Little

1987. “Industrial Technology Development: A Network Approach, Edited by Håkan Håkansson.” Journal of Product Innovation Management 4 (2): 163–65.

60.

Mancusi

M. L.

2008. “International Spillovers and Absorptive Capacity: A Cross-Country Cross-Sector Analysis Based on Patents and Citations.” Journal of International Economics 76 (2): 155–65.

61.

Mkono

Tribe

. 2016. “Beyond Reviewing: Uncovering the Multiple Roles of Tourism Social Media Users.” Journal of Travel Research 56 (3): 287–98.

62.

Moon

Chadee

Tikoo

. 2008. “Culture, Product Type, and Price Influences on Consumer Purchase Intention to Buy Personalized Products Online.” Journal of Business Research 61 (1): 31–39.

63.

Mowery

Rosenberg

. 1979. “The Influence of Market Demand upon Innovation: A Critical Review of Some Recent Empirical Studies.” Research Policy 8 (2): 102–53.

64.

Murphy

P. E.

2013. Tourism: A Community Approach (RLE Tourism). Oxon: Routledge.

65.

Ness

Aarstad

Haugland

S. A

. 2014. “Destination Development: The Role of Interdestination Bridge Ties.” Journal of Travel Research 53 (2): 183–95.

66.

Ning

Wang

. 2016. “Urban Innovation, Regional Externalities of Foreign Direct Investment and Industrial Agglomeration: Evidence from Chinese Cities.” Research Policy 45 (4): 830–43.

67.

Nooteboom

2000. “Learning by Interaction: Absorptive Capacity, Cognitive Distance and Governance.” Journal of Management and Governance 4(1/2): 69–92.

68.

Ottaviano

Peri

. 2006. “The Economic Value of Cultural Diversity: Evidence from US Cities.” Journal of Economic Geography 6 (1): 9–44.

69.

Ozgen

Nijkamp

Poot

. 2012. “Immigration and Innovation in European Regions.” In Migration Impact Assessment: New Horizons, edited by Nijkamp

Poot

J. Y.

Sahin

, 261–300. Cheltenham, UK: Edward Elgar.

70.

Piperopoulos

Wang

. 2017. “Outward FDI, Location Choices and Innovation Performance of Emerging Market Enterprises.” Research Policy 47 (1): 232–40.

71.

Pot

Vaas

. 2008. “Social Innovation, the New Challenge for Europe.” International Journal of Productivity and Performance Management 57 (6): 468–73.

72.

Putnam

R. D.

2000. Bowling Alone: The Challenge and Revival of American Community. New York: Simon & Schuster.

73.

Quigley

J. M.

1998. “Urban Diversity and Economic Growth.” Journal of Economic Perspectives 12 (2): 127–38.

74.

Romão

Nijkamp

. 2017. “Impacts of Innovation, Productivity and Specialization on Tourism Competitiveness—A Spatial Econometric Analysis on European Regions.” Current Issues in Tourism 2:1–20.

75.

Saha

J. J.

Campbell

. 2016. “Does Political and Economic Freedom Matter for Inbound Tourism? A Cross-National Panel Data Estimation.” Journal of Travel Research 56 (2): 221–34.

76.

Schmookler

1966. Invention and Economic Growth. Cambridge, MA: Harvard University Press.

77.

Smith

2009. Issues in Cultural Tourism Studies. Oxon: Routledge.

78.

Solow

R. M.

1957. “Technical Change and the Aggregate Production Function.” Review of Economics and Statistics 39 (3): 312–20.

79.

State Intellectual Property Office of China. 2015. Annual Report of Valid Patents in China of 2014.

80.

Tian

Wang

H. H.

Chen

. 2010. “Spatial Externalities in China Regional Economic Growth.” China Economic Review 21:S20–31.

81.

Tisdell

Wen

. 1991. “Foreign Tourism as an Element in PR China’s Economic Development Strategy.” Tourism Management 12 (1): 55–67.

82.

Trippl

2010. “Developing Cross-Border Regional Innovation Systems: Key Factors and Challenges.” Tijdschrift voor Economische en Sociale Geografie 101 (2): 150–60.

83.

Verdu-Jover

A. J.

Alos-Simo

Gomez-Gras

J. M

. 2017. “Adaptive Culture and Product/Service Innovation Outcomes.” European Management Journal. Published online July 17. http://doi.org/10.1016/j.emj.2017.07.004.

84.

Vita

G. D.

Kyaw

K. S

. 2016. “Tourism Specialization, Absorptive Capacity and Economic Growth.” Journal of Travel Research 56 (4): 423–35.

85.

Y. M.

2006. “A Spatial Econometric Model and Its Application to Research & Development and Regional Innovation.” Journal of Quantitative & Technical Economics 5:74–85.

86.

Wilson

1997. “Paradoxes of Tourism in Goa.” Annals of Tourism Research 24 (1): 52–75.

87.

Yang

Wong

K. K. F

. 2012. “A Spatial Econometric Approach to Model Spillover Effects in Tourism Flows.” Journal of Travel Research 51 (6): 768–78.

88.

Yoon

Yun

Lee

Phillips

. 2015. “Entrepreneurship in East Asian Regional Innovation Systems: Role of Social Capital.” Technological Forecasting and Social Change 100:83–95.

89.

Zach

F. J.

Hill

T. L

. 2017. “Network, Knowledge and Relationship Impacts on Innovation in Tourism Destinations.” Tourism Management 62:196–207.

90.

Zahra

A. Z.

George

. 2002. “Absorptive Capacity: A Review, Reconceptualization, and Extension.” Academy of Management Review 27 (2): 185–203.