Sage Journals: Discover world-class research

Abstract

Using a bilateral dataset of trips for 32 European countries for the period 2012–2019, we aim to gauge the size and evolution of the border effect on tourism: the extent to which domestic tourism is greater (or lesser) than international tourism. We found that the tourism flow within countries was 24 times greater than between countries. Also, we show that, relative to 2012, the border effect diminished by 13% in 2019. Our results suggest that the size (and evolution) of the border effect is the same for trips that last between one and three nights (short trips) and those that last four nights or more (long trips). Nonetheless, our findings show that bilateral determinants that represent travel and transaction costs are more important when explaining short trips than long ones.

Keywords

border international tourism Europe gravity

Introduction

International tourism is a consolidated, rapidly expanding industry; since the 1950s tourism activities have increased by an average of 6% per year (Eriksen, 2016). It generates more than half a trillion dollars in annual revenues and represents 10% of total international trade and almost half of total trade in services (Eilat and Einav, 2004). Considering the global aggregate, tourism contributes, directly or indirectly, to one in 10 jobs, and is responsible for 10.3% of the global GDP (WTTC, 2020).

International mobility and borders are two inseparable concepts of analysis (Richardson, 2013), since, as pointed out by Rumford (2006: 155) “to theorize mobilities and networks is at the same time to theorize borders”. Tourists are a particular type of mobile subject that crosses a world full of borders, their movements affect and are affected by the construction and performance of these borders (Rowen, 2016). Certainly, there is a vast academic literature that relates mobility to the performance of state borders (Parsley 2003; Dauvergne 2004; Salter 2006, 2008; Wonders 2006). Whereas the study of tourism flows in relation to the effect of borders has received very little attention (Rowen, 2016). Some studies have been published on the role of borders in encouraging or restricting tourism (Timothy, 2005; Timothy and Tosun, 2003; Sofield, 2006; Tubadji and Nijkamp, 2018), but the literature that quantifies the extent to which national borders condition tourism flows is very limited. This present paper seeks to fill this gap in the literature: to quantify the size and evolution of the border effect on international tourism.

From a theoretical point of view, the analysis of the impact of borders on international economic relations is simultaneous with the development of the economy as a discipline. However, empirical estimates of the border effect are relatively recent. The basic idea that gives meaning to the analysis of the border effect derives from the fact that borders represent a legal, physical, cultural and psychological barrier that hinders the exchange of goods, services, capital and people between countries, so that these exchanges will always be - everything else being equal - more fluid between regions in the same country than between territories in different countries. As in other areas of international economy, the border effect began to be estimated for the trade in merchandise (McCallum, 1995), and was subsequently extended to other areas of international economic relations, such as trade in services (Anderson et al., 2018), foreign investment (Mayer et al., 2010), tourism (Paniagua et al., 2022) and forced migrations (Carril-Caccia et al., 2021).

This paper proposes a country-level study focused on a sample of European countries. The selection of this area is justified considering that, in terms of tourism, it forms a space for the free movement of people that is unique in the world. Therefore, legal and administrative restrictions are completely removed from the equation. Europe is a recognized tourism power, accounting for half of international tourism flows (UNWTO, 2019). For a large number of member states of the European Union (EU), tourism has been a major driver of economic growth and job creation (Salinas et al., 2020). Tourism contributes not only to the economic growth of countries and regions, but also to their social and cultural development and their general well-being (Martín et al., 2020). In particular, in the EU, borders have become, with regard to mobility and trade, symbolic representations of integration and cooperation (Więckowski, 2010). In addition, the implementation of the monetary union has broken down the intangible border associated with the different currencies in circulation, something that also drives tourist movements (Khalid et al., 2021).

To the best of our knowledge, Anderson et al. (2018) and Paniagua et al. (2022) are the only other works that have attempted to quantify the border effect on tourism. The present analysis contributes to the literature in four different ways. First, we focus on a sample of countries for which, during our period of analysis, there were no formal barriers to touristic flows. Second, we do not approximate domestic tourism; we use statistics that directly measure domestic and international tourism¹. Third, the present work also contributes to the literature by quantifying the evolution of the border effect. Fourth, our analysis sheds new light on the different drivers of tourism depending on the length of the stay.

In the present work, for a sample of European countries, we estimate a structural gravity model with domestic and international tourism flows. Our analysis shows that domestic tourism is 24 times greater than international. In addition, our estimates indicate that the border effect is lower for long trips (trips that last four nights or more) than for short trips (trips that last between one and three nights), although the difference in size is not statistically significant. Furthermore, we demonstrate that between 2012 and 2019 for both, long and short trips, the border effect diminished by nearly 13%. In addition, our findings suggest that travel costs and transaction costs have a significantly greater impact on short trips than on long ones. Our results remain qualitatively and quantitatively unchanged when the sample is restricted to EU countries.

The following section provides an overview of the literature related to the drivers of tourism flows and the border effect. The Methodology and data section presents the data and empirical strategy, and the Results section reports the results. The article ends with a discussion of the results and some concluding remarks.

The border effect on touristic flows

Historically, international touristic flows have increased as transport technology has evolved, but at the same time they have been hampered by the territorial rights exercised by sovereign countries through frontiers (Sofield, 2006). The generally accepted definition of international tourism includes the process of crossing the border from one country to another, staying at least one night (UNWTO, 1991), as well as the recreational and temporary nature of the trip (Hall, 2005).

The study of international flows and the border effect has experienced significant change in the last decades, as a result of international trade and flows of people and knowledge growing smoother and altering the traditional role of borders (Hageman et al., 2004). There are many academic papers on the implications for mobility of a reshaped world map resulting from decades of decolonization, territorial changes following the Cold War and the collapse of the Soviet Union, and regional integration processes. However, only a few studies specifically examine the role of borders on tourism. Some exceptions are the research conducted by Timothy (1999, 2001, 2005), Więckowski and Timothy (2021), Timothy and Tosun (2003), several of which have a qualitative nature.

The behaviour of tourists during their trips cannot be separated from the geographical, cultural and political conditioning factors on a national scale (Rowen, 2016). Borders are understood to be institutions that mark a functional barrier between countries, impose control over flows of people and trade, identify the entry point and facilitate contact and exchange (Hageman et al., 2004). International borders have geopolitical, historical and symbolic values and meanings (Więckowski and Timothy, 2021) which have an influence on tourists when choosing a destination. In a globalised world, the choice of a tourist destination is determined by a great variety of socioeconomic and environmental factors. This makes demand difficult to predict and it complicates tourism management (Albrecht, 2013; Liu et al., 2015; Song et al., 2017; Van der Zee and Vanneste, 2015; Von Bergner and Lohmann, 2014). The previously mentioned border effect can be produced in tourism by two types of factors: the administrative and physical elements that limit tourist movement and the psychological and perception elements that generate the image of crossing to another reality or into the unknown (Timothy and Tosun, 2003). This means that regions located in neighbouring countries have a lower level of interaction than expected (Smith, 1984; Timothy and Tosun, 2003).

Timothy and Tosun (2003) analysed touristic flows at the border between the US and Canada. They concluded that borders can create real barriers that can be perceived by the traveller (even between countries with strong bonds of friendship and cooperation) due to the existence of economic and cultural differences, when there are security problems or when crossing procedures are tedious for the visitor. Sofield (2006) focused on border tourism and on the factors that can revitalize the roles of borders (such as international security problems and pandemics). More recently, Mayer et al. (2019) focused on this topic, whereas the attraction of travelling is, for some tourists, the fact of crossing a border and being immersed in a foreign culture, for others, it could be a bother and cause anxiety. Therefore, as in the case of the previous literature on trade in merchandise and services, FDI and forced migration (e.g. Anderson et al., 2018; Carril-Caccia et al., 2021; Umber et al., 2014), we expect touristic flows to have a positive border effect. That is to say, after controlling for multiple drivers of international tourism, we expect domestic touristic flows to be greater than international ones. In addition, in line with previous works focused on international trade (e.g. Anderson et al., 2018; Bergstrand et al., 2015; Franco-Bedoya and Frohm, 2021), as a result of the advance of globalization, we expect to find a reduction in the border effect over time.

The hypotheses we posit are to a certain extent aligned with the previous findings by Anderson et al. (2018) and Paniagua et al. (2022). The first finds that the border effect has diminished for international trade in services, but also finds that for trade in travel services the border effect is negative. That is to say, after controlling for multiple drivers of international tourism, Anderson et al. (2018) find that international tourism is greater than domestic. Paniagua et al. (2022) find a positive border effect. Also, for two of the three proxies that domestic tourism authors’ employ, they obtain results that suggest a reduction in the border effect. However, when Paniagua et al. (2022) proxy domestic tourism with the number of guests, their results suggest an increase in the border effect.

Methodology and data

Methodology

In the mid-1990s, taking the gravity models that had been used in economics since the 1960s as a basis -which in their simplest version consider that exchanges are directly proportional to the economic size of the countries and inversely proportional to the physical distance that separates them-the first estimations of the border effect were made for the Provinces of Canada and the States of the USA (McCallum, 1995). The measurement was made by introducing a dummy in the gravity equation, which takes the value 0 for regions within the same country and one when the regions are separated by a border. Anderson and Van Wincoop (2003) criticized the work of McCallum (1995) for being based on a gravity equation without theoretical foundations, which could introduce important biases in the estimates due to specification errors. Starting from the theoretical formulations of Anderson (1979) and Bergstrand (1985), Anderson and Van Wincoop (2003) developed a new gravity model with theoretical foundations and tested it empirically. Bergstrand et al. (2015) set out to simultaneously measure the impact of trade integration agreements, borders, and physical distance on trade flows, concluding that previous estimates were biased upwards.

We use the gravity equation to estimate the border effect on tourism. Mimicking Newton’s universal law of gravitation, the gravity model for tourism predicts that international travel will be positively moderated by the countries’ economic masses (or population) and negatively by the geographic distance between them. This model has been widely and successfully used by previous literature that sought to explain the drivers of bilateral tourism (e.g. Gil-Pareja et al., 2007; Keum, 2010; Santana-Gallego and Paniagua, 2022). In addition, Morley et al. (2014) provided a theoretical background that supports the use of the gravity equation to model tourism demand. We estimate the following equation:

{T o u r i s m}_{i j t} = e^{(\begin{array}{c} α {B o r d e r}_{i j} + {β_{1} D i s t a n c e}_{i j} + {β_{2} C o n t i g u i t y}_{i j} + β_{3} {R e l i g i o n}_{i j} + β_{4} {L a n g u a g e}_{i j} \\ {μ_{1} E M U}_{i j t} + μ_{2} {M i g r a t i o n}_{i j t - 3} + λ_{i t} + λ_{j t} \end{array})} * ε_{i j t}

(1)

where

{T o u r i s m}_{i j t}

is the number of tourism trips from country

i

to country

j

in year

t

. The dependent variable includes domestic

(i \neq j)

and international

(i \neq j)

trips.

{B o r d e r}_{i j}

is a dummy variable that takes the value one when

i

and

j

are the same country, and 0 when the trips are international

(i \neq j)

. Accordingly, the coefficient

α

gauges the extent to which domestic tourism is greater (or lesser) than international. The equation also includes bilateral time-invariant determinants of tourism (e.g. Khadaroo and Seetanah, 2008; Patuelli et al., 2013; Tubadji and Nijkamp, 2018); the logarithm of geographic distance, contiguity and language, which are indicator variables that respectively have a value of one if a country pair share a geographic border and language, and a religion similarity index.

Border and Contiguity are different explanatory variables. The first quantifies the relevance of internal flows relative to international ones, while Contiguity represents the effect that sharing a geographic border with another country has on international flows. Accordingly, the coefficient associated with Border $(α)$ quantifies the extent to which domestic tourist trips are longer (or shorter) than international ones. In other words, the preference and/or capacity that tourists have for crossing their country’s border to travel to a foreign country. The coefficient associated with the Contiguity dummy $(β_{2})$ assesses the degree to which sharing a geographic border between two countries, such as Portugal and Spain, affects bilateral tourism between them.

${E M U}_{i j t}$ is an indicator variable that has a value of one if countries are members of the European Monetary Union (EMU). Sharing a currency can benefit tourism as it eliminates exchange rate volatility, reduces transaction costs and improves the transparency of markets. In summary, the adoption of a common currency between a pair of countries can reduce the cost of tourism between them. In this regard, Gil-Pareja et al. (2007) and Santana-Gallego et al. (2016) demonstrate that the adoption of the Euro had a positive effect on tourism flows between EMU members.

${M i g r a t i o n}_{i j t - 3}$ is the logarithm of the population of migrants from country $i$ that live in country $j$ in the year $t - 3$ . Santana-Gallego and Paniagua (2022) argue that migration can foster tourism through three different channels. First, migrants can reduce the bilateral information asymmetries that might exist between their country of origin and their host country, in this way making it easier to travel from one country to another. Second, migrants can also directly and indirectly diminish the costs associated with travelling. They can directly reduce the costs of travelling by offering cheaper or free accommodation to their compatriots, and indirectly as they intensify transport connections between a pair of countries and in this way reduce transport costs. Third, migrants can promote their new home country in their country of origin as an attractive destination to visit.² The theoretical framework and empirical evidence presented by Santana-Gallego and Paniagua (2022) support this positive link between tourism and migration.³

Since our analysis covers the 2012–2019 period for a sample of highly integrated European countries, we do not need to control for other relevant time variant bilateral drivers of tourism such as trade agreements or the cost of visas (Balli et al., 2013; Santana-Gallego et al., 2016).

Equation (1) includes origin country-year fixed effects $(λ_{i t})$ and destination country-year fixed effects $(λ_{j t})$ . Both respectively absorb all origin and destination country level time-varying drivers of tourism flows identified by the previous literature, e.g. population, GDP per capita, happiness, accommodation capacity, violence, institutional quality, and price (e.g. Balli et al., 2016; Dogru et al., 2017; Llorca-Vivero, 2008; Paniagua et al., 2022; Xu et al., 2019). Thus, country specific time varying variables, such as GDP, are not included in the gravity equation as they are collinear with the origin country-year fixed effects and destination country-year fixed effects. More importantly, these fixed effects account for the multilateral resistance term (MRT) of the origin and destination countries (Anderson and Van Wincoop, 2003; Fally, 2015). This implies controlling for the remoteness of two country pairs to other possible destinations. That is to say, the MRT controls for the relative capacity of travelling from a given country in comparison to the rest of the world. Likewise, the destination countries’ MRT controls for the relative capacity of a country to attract tourists relative to the rest of the world (Harb and Bassil, 2020; Santana-Gallego and Paniagua, 2022). Not controlling for the MRT results in biased estimates (Head and Mayer, 2014). Finally, $ε_{i j t}$ is the disturbance term.

In this paper the concept of distance is addressed from a spatial point of view, although there are other non-physical factors that increase the perception of distance between two countries (Popli et al., 2016). These factors correspond to social and cultural differences that can modify human and economic interactions between territories (Björkman et al., 2007; Basuil and Datta, 2015), acting as a mental barrier (Maddux and Galinsky, 2009). Hence, the importance of including elements such as language, social values, cultural aspects, religion and contiguity in the analysis, which in a way capture distances/cultural frontiers, as this paper proposes. In the European context, there are several forces that condition - in different ways - the non-physical and non-political barriers between countries. Such as the creation of a European identity and policies that encourage interaction between citizens and companies, the reinforcement of national identity and the homogenization processes associated with globalization (Weidenfeld, 2013).

Equation (1) is estimated to gauge the size of the border effect. Equation (2) measures the evolution of the border effect during our period of analysis:

{T o u r i s m}_{i j t} = e^{(α_{t} {B o r d e r}_{i j} * D_{t} + μ_{1} {E M U}_{i j t} + μ_{2} {M i g r a t i o n}_{i j t - 3} + λ_{i j} + λ_{i t} + λ_{j t})} * ε_{i j t}

(2)

As in Bergstrand et al. (2015) the ${B o r d e r}_{i j}$ dummy interacts with a dummy variable that takes a value of one when the observation is from year $(D_{t})$ . The estimated coefficient $α_{t}$ measures the change in the border effect in year $t$ relative to the base year. This empirical strategy has also been adopted in other works that address the evolution of the border effect for trade, migration and FDI (Anderson et al., 2018; Bergstrand et al., 2015; Franco-Bedoya and Frohm, 2021; Head and Mayer, 2021). In equation (2), bilateral time-invariant determinants of tourism are replaced with country-pair fixed effects $(λ_{i j})$ . These fixed effects reduce the omitted variable bias and the endogeneity that there is between tourism flows and sharing a currency (Baldwin and Taglioni, 2006; Santana-Gallego et al., 2016). In consequence, they allow us to better approximate the change in the border effect over time. Nonetheless, country-pair fixed effects are collinear with the ${B o r d e r}_{i j}$ dummy, thus equation (2) does not provide any insight regarding the size of the border effect.

Equations (1) and (2) are estimated with the Pseudo Poisson Maximum Likelihood (PPML) estimation. Since the work from Santos-Silva and Tenreyro (2006), the PPML has been the standard estimator for the gravity model. The estimator allows estimation of a log linear model at the same time that we are able to conserve the zeros usually present in bilateral data and to overcome the heteroskedasticity problems usually present with Ordinary Least Squares. Following Egger and Tarlea (2015), robust standard errors are multiway clustered at the origin and destination country. We use the PPML high-dimensional fixed effects estimator from Correia et al. (2020).

Data overview

The data about international and domestic tourism is from Eurostat. We used a bilateral dataset of the number of tourists between 29 source and 32 destination European countries for the period 2012–2019 (The list of countries and details about the treatment of the tourism data are available in the Supplemental Appendix). With the Eurostat data we can have separate statistics for the number of tourists according to different lengths of stay: the number of tourists that stayed between one and three nights and the number that stayed four or more nights. A limitation of our data is that it does not account for tourists spending their trip in establishments that are outside the market (e.g. a family house). This limitation is likely to be present in both, international and domestic trips, however it might be more important in the case of domestic trips. Nonetheless, our database should be useful to compare domestic and international tourism that is measured in a homogeneous way.

Figure 1 illustrates the evolution of the ratio of international-domestic tourists by different lengths of stay. As expected, independent of the length of stay the number of domestic tourists is significantly larger than the number of international ones. Nevertheless, the difference is smaller for tourists that stay four or more nights. On average, for each one hundred domestic tourists that stay one to three nights, there are nearly 15 that holiday abroad. For each one hundred domestic tourists that stay four nights or more, nearly 57 stay abroad. In both cases there appears to be a positive trend in the ratio of international-domestic tourists, growth being more marked for the case of tourist trips that last four nights or more. Alternatively, Table 1 shows that on average the number of tourists that stay between one and three nights is larger than the number of tourists that stay four nights or more. Short trips are more frequent and more likely to occur domestically than longer ones.

Figure 1.

Ratio of international-domestic tourists. Source: The Authors. Data from Eurostat.

Table 1.

Descriptive statistics.

Variable	Obs	Mean	Std. Dev.	Min	Max
No. Tourists: 1 or more nights	5483	1526.795	11,234.93	0.175	200,379.7
No. Tourists: 1–3 nights	5401	925.8511	7365.677	0.001	114,649.5
No. Tourists: 4 or more nights	5401	622.608	4106.766	0.031	86,881.88
Border	5483	0.042	0.200	0	1
Log(Distance)	5483	6.946	0.752	2.134	8.286
Contiguity	5483	0.122	0.327	0	1
Religion	5483	0.319	0.303	0	0.956
Language	5483	0.054	0.227	0	1
EMU	5483	0.350	0.477	0	1
Log(Migration)	5483	6.787	4.147	0	18.226

Source: the authors. The number (No.) of tourists is reported in thousands.

The variables for geographic distance, contiguity, religion similarity and language were obtained from the CEPII database (Head et al., 2010). Geographic distance between countries is the population-weighted distance between the most populated cities measured in kilometres. Contiguity is an indicator variable that has the value one when two countries share a geographic border. The religion similarity index was obtained by adding the products of the proportions of the different religions in the origin and destination countries. It is an index that can have a value between 0 and 1, the higher the value of the index, the higher the religious similarity between a pair of countries. Language is a dummy variable that has the value one when two countries share the same official or principal language. The EMU variable was constructed by the authors. This variable has a value of one when two countries are members of the EMU. Data on the stock of international migrants and domestic non-migrant population was obtained from Eurostat. For some countries, missing data on the domestic non-migrant population was imputed using a population data net of international migrants obtained from the World Bank’s Development Indicators⁴. Descriptive statistics are available in Table 1.

Results

Table 2 presents equations (1) and (2), regressions for the number of tourist trips of one night or more. For both, we estimate for the whole sample (columns 1 and 3) and a restricted sample that only includes EU countries (columns 2 and 4). Estimates in column 1 show a negative relationship between geographic distance and tourist trips, while religious affinity and sharing a language have a positive significant effect. Sharing a geographic border has a non-significant effect. Results for the EU sample are similar; the only difference is that religious affinity looses significance.

Table 2.

The border effect on tourism trips of one night or more.

	(1)	(2)	(3)	(4)
	Equation (1)	Equation (1)-EU	Equation (2)	Equation (2)-EU
Border	3.1983^a	3.1396^a
	(0.386)	(0.338)
Log(Distance)	−0.9784^a	−1.0577^a
	(0.179)	(0.174)
Contiguity	0.0901	0.0437
	(0.160)	(0.162)
Religion	0.4825^c	0.3869
	(0.259)	(0.263)
Language	1.2776^a	1.3410^a
	(0.108)	(0.155)
EMU	0.4050^a	0.4546^a	0.0906^b	0.0989^a
	(0.084)	(0.095)	(0.039)	(0.038)
Log(Migrant population t-3)	−0.0035	−0.0034	0.1134^b	0.1040^c
	(0.031)	(0.028)	(0.050)	(0.053)
Observations	5483	4539	5450	4516
Origin-year FE	Yes	Yes	Yes	Yes
Destination-year FE	Yes	Yes	Yes	Yes
Country pair FE			Yes	Yes
Border-year			Yes	Yes

Note: Standard errors clustered at origin and destination level are in parentheses; a, b, c: statistically significant at 1%, 5%, and 10%, respectively. Estimates from the border-year dummies are reported in Figure 2 and Figure A in the Supplemental Appendix.

Columns 3 and 4 show that being a member of the EMU has a significant, positive effect on the number of tourism trips. ⁵ The positive effect of the EMU on the number of tourists trips is aligned with the previous literature that shows the positive effect that the EMU has between country members on tourism, international trade, cross-border M&As, bonds and equity holdings (e.g. Coeurdacier and Martin, 2009; Coeurdacier et al., 2009; Gil-Pareja et al., 2007; Glick and Rose, 2016).

Estimates in columns 3 and 4 indicate that the stock of migrant population from country $i$ in country $j$ is positively correlated to the number of tourist trips. This positive, significant link between tourism and migration is sensitive to the number of years in which we lag the migration stock variable. The variable remains positive and significant in $t - 2$ , but loses significance when it is included in the model in years $t$ , $t - 1$ and $t - 4$ . Estimates of the remaining independent variables remained identical to the ones presented in table 2. To conserve space, this robustness analysis is available on request.

Regarding the border effect, columns 1 and 2 from Table 2 corroborate that even in the case of analyzing a sample of highly integrated countries, there was a stronger preference for travelling domestically than internationally. Our estimates show that domestic trips were between 23 $(e^{3.1396})$ and 24.5 $(e^{3.1983})$ times greater than international ones. Based on this, we did not find that the border effect was statistically different between the full sample and the sample restricted to EU countries.⁶

Figure 2 illustrates the evolution of the border effect relative to the base year 2012. These estimates were obtained from equation (2) (column 3). The results show that during the years 2014–2016 there was no significant change in the border effect. However, in years 2018 and 2019 there was a reduction in the border effect of 11% $(e^{- 0.1170} * 100)$ and 12.8% $(e^{- 0.1368} * 100)$ respectively relative to 2012. As in the case of the border effect, we found that there were no significant differences between the evolution of the border effect of the whole and EU sample (to conserve space, there is a Figure in the Supplemental Appendix that illustrates the evolution of the border effect of both samples).

Figure 2.

Evolution of the border effect on tourism trips of one night or more. Note: Evolution of the border effect, 2012–2019. Change of the border effect relative to the border effect in 2012. Coefficients are obtained from estimating equation (2) with the whole sample.

Table 3 shows the estimates of equations (1) and (2) for the number of tourist trips of one to three nights (columns 1 and 2) and for the number of tourist trips of four nights or more (columns 3 and 4). Depending on the length of the stay, the results show that there were some statistically significant differences between the drivers of tourist trips. For short stays, one to three nights, the negative effect of geographic distance was almost three times greater than for longer stays (−1.5036 vs −0.6370). In this line, for shorter stays being an EMU member, and sharing a common geographic border and religious affinity are significant drivers of tourist trips, while they are not significant in the case of longer stays. In addition, having a common language has a greater positive effect for one to three night trips (1.5257 vs 1.0794). In this way, the results seem to indicate that for short trips, tourists prefer to travel to places which have lower travel and transaction costs. On the other hand, the migrant population is only positive and significant when considering the number of four night or more trips.

Table 3.

The border effect on tourism trips of one to three nights and four or more nights.

	(1)	(2)	(3)	(4)
	One to three nights		Four or more nights
	Equation (1)	Equation (2)	Equation (1)	Equation (2)
Border	3.1849^a		2.9924^a
	(0.439)		(0.414)
Log(Distance)	−1.5036^a		−0.6367^a
	(0.202)		(0.193)
Contiguity	0.2535^b		0.0203
	(0.113)		(0.167)
Religion	1.0642^a		0.1636
	(0.251)		(0.279)
Language	1.5257^a		1.0794^a
	(0.118)		(0.133)
EMU	0.2954^a	0.0817^b	0.4347^a	0.0966
	(0.114)	(0.037)	(0.099)	(0.067)
Log(Migrant population t-3)	−0.0151	0.1364	0.0068	0.1050^b
	(0.023)	(0.104)	(0.040)	(0.044)
Observations	5401	5396	5401	5396
Origin-year FE	Yes	Yes	Yes	Yes
Destination-year FE	Yes	Yes	Yes	Yes
Country pair FE		Yes		Yes
Border-year		Yes		Yes

In terms of the border effect, estimates from Table 3 are to a certain extent aligned with the descriptive statistics reported in Figure 1. The preference for making domestic tourists trips was greater for short trips than for long trips. For trips of one to three nights, estimates show that domestic trips were 24.2 $(e^{3.1849})$ times greater than international ones. For trips of four nights or more, estimates indicated that domestic trips were 19.9 $(e^{2.9924})$ times greater than international ones. Nonetheless, the difference in the border effect size was not statistically significant.

Figure 3 presents the evolution of the border effect relative to the base year 2012 for those trips that last between one and three nights and those that last four nights or more. Estimates were obtained from columns 2 and 4 respectively. In both cases, there was a reduction of the border effect, in 2019 they were 10.9% $(e^{- 0.1156} * 100)$ and 13.4% $(e^{- 0.1435} * 100)$ lower respectively than in 2012. However, the estimates show that there were no significant differences in the evolution of the border effect between both.

Figure 3.

Evolution of the border effect on tourism trips of one to three nights and four nights or more. Note: Evolution of the border effect, 2012–2019. Change of the border effect relative to the border effect in 2012. Coefficients are obtained from estimating equation (2) with the whole sample.

Discussion and conclusions

The study of the factors that influence international tourism flows is an object of increasing attention. In the past two decades, a lot of empirical works have emerged; most of them based on gravity models, which try to explain the general patterns observed in tourist movement on a global scale. This research is intended to contribute to that existing literature. It presents a gravity model with domestic and international tourist flows for a sample of 32 European countries for the period 2012–2019. Generally, referring to the most conventional hypothesis, our results are in concordance with the previous literature. Tourist flows are sensitive to physical distance, religious and linguistic affinity, the fact of sharing the same currency, and the previous existence of a certain stock of migrant population.

Regarding our most novel hypothesis, the results indicate that the border effect is a factor that greatly influences the travel decisions of tourists. Our results prove that domestic tourism is between 19 and 24 times higher than international tourism. Furthermore, in consonance with globalization, it was found that there was a reduction in the border effect between 2012 and 2019. This means that the tendency of the border effect to diminish could probably also be explained by the very same reasons that explain its loss of significance in commercial exchanges (reduction relating to international transport cost, opening of borders in the globalization scenario, etc.).

In addition, considering the type of trip and as economic intuition suggests, the negative effect of geographic distance on shorter trips is greater than on longer trips, while sharing a currency or language, having a common border, and having a religious affinity might have a greater positive effect on shorter trips. As far as the migrant population is concerned, it only seems to have a positive influence on long trips. Stated briefly, our results indicate that tourists prefer to incur lower travel and transaction costs when making short trips. However, we find no evidence that the border effect differs according to trip length.

These results could be relevant when it comes to designing tourism promotion policies by governments, and defining business strategies in order to attract travellers. The variables influencing short and long trips are significantly different in each case, and the source markets of tourists may necessarily be oriented towards one or the other travel mode, as is the case, for example, with transatlantic trips, which logically need to be scheduled for more days. The lure of a common language and culture may be relevant in certain markets or types of travel, but be much less relevant in other cases. Promotional campaigns for specific dates (weekends, special events such as fairs, exhibitions, etc.) may make sense in certain countries or types of travel, but not in others. In conclusion, policy makers and tourism businesses need to be aware of the differences in the factors influencing different types of travel in order to develop effective and efficient policies and business strategies.

Considering previous literature, we found that the border effect is close to the lower rank of the estimations given by Paniagua et al. (2022). The aforesaid work found that the border effect swings in such a manner that residential tourism flows are between 19 and 117 times greater than international ones. The rank of estimations of Paniagua et al. (2022) is determined by the different strategies used to approach residential tourism. In other matters, as opposed to Anderson et al. (2018) and the travel trade, we can see a positive frontier effect.

Furthermore, our estimates indicate that the border effect on tourism is greater than for international trade and FDI, but lower than for forced migration. Bailey et al. (2021) show that domestic trade is 5–9 times greater than international. Umber et al. (2014) conclude that domestic M&A is 6 times greater than cross-border. Carril-Caccia et al. (2021) found that domestic forced migration is 42 times greater than international. This is probably due to the greater subjective effect that frontiers have on personal decisions regarding travelling. Therefore, aligned with the previous literature on trade and forced migration (Anderson et al., 2018; Bergstrand et al., 2015; Carril-Caccia et al., 2021; Franco-Bedoya and Frohm, 2021; Head and Mayer, 2021), we also observe a decline over time in the border effect on tourism.

Finally, we should highlight that this work focuses on a geographic field and a quite homogeneous period of time from an economic, political, health and safety point of view. This homogeneity, most likely, is behind the strength of the obtained results. Nonetheless, at the same time, it constitutes a limitation of our analysis, due to the lack of diversity that stops our models from capturing more precisely the effects of cultural differences, administrative constraints, migratory flows, pandemics, the political situation and global security at a global scale. All of this indicates that we should increase the sample of countries and the analysed period of time in future works, contemplating them from an intercontinental perspective and covering more heterogeneous periods. Moreover, future research could address different dimensions that can potentially moderate the border effect. For instance, the border effect could be moderated by the cultural differences or social ties that there are between country pairs. In addition, country level factors, such as institutional quality and the number of World Heritage sites, could also affect the size of the border effect.

Supplemental Material

Supplemental Material - How important are borders for tourism? The case of Europe

Supplemental Material for How important are borders for tourism? The case of Europe by Federico Carril-Caccia, José María Martín Martín, Francisco Javier Sáez-Fernández in Tourism Economics

Footnotes

Acknowledgment

We are very grateful for the feedback from two anonymous reviewers. This research was conducted as part of the Project PID2021-122133NB-I00 financed by MCIN/AEI/10.13039/501100011033/FEDER, EU. We also gratefully acknowledge financial support from the Junta de Andalucía (SEJ 413) and Generalitat Valenciana (CIAICO/2021/006).

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) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

José María Martín Martín

Supplemental Material

Supplemental Material for this article is available online.

Notes

Author biographies

Federico Carril-Caccia is an assistant professor from the Department of Spanish and International Economics at the University of Granada (Spain). He obtained his PhD in Economics from University of Granada, in 2019. His research focuses on the drivers of tourism and migration, and the determinants and effects of foreign direct investment. His work has been published in various international journals such as International Journal of Emerging Markets, World Economy, and Empirical Economics.

José Maria Martín is senior lecturer at University of Granada, Spain, Department of Spanish and International Economics. The interests of his research focus on tourism economics, economic and social sustainability, seasonality and sustainable development. He has worked on numerous consulting projects for public institutions and private companies. He has held various academic management positions at the University of Granada and the International University of La Rioja, as well as in various consulting companies.

Francisco Javier Sáez-Fernández is Associate Professor of Applied Economics at the University of Granada. His research work focuses on the fields of Banking, Tourism Economics and Water Resources Management. He has published about a hundred works, including books, book chapters and scientific papers. He has directed almost twenty research projects and contracts, both public and private. He has been a visiting professor at numerous foreign universities, including Harvard University, University College London and London School of Economics. He is currently Head of the Department of International and Spanish Economics

References

Albrecht

(2013) Networking for sustainable tourism— towards a research agenda. Journal of Sustainable Tourism 21(5): 639–657.

Anderson

(1979) A theoretical foundation for the gravity equation. The American Economic Review 69(1): 106–116.

Anderson

Borchert

Mattoo

, et al. (2018) Dark costs, missing data: shedding some light on services trade. European Economic Review 105: 193–214.

Anderson

van Wincoop

(2003) Gravity with gravitas: a solution to the border puzzle. The American Economic Review 93(1): 170–192.

Bailey

Gupta

Hillenbrand

, et al. (2021) International trade and social connectedness. Journal of International Economics 129: 103418.

Baldwin

Taglioni

(2006) Gravity for Dummies and Dummies for Gravity Equations. National Bureau of Economic Research. Working Paper 12516.

Balli

Cebeci

(2013) Impacts of exported Turkish soap operas and visa-free entry on inbound tourism to Turkey. Tourism Management 37: 186–192.

Balli

Rosmy

(2016) The impacts of immigrants and institutions on bilateral tourism flows. Tourism Management 52: 221–229.

Basuil

Datta

(2015) Effects of industry- and region-specific acquisition experience on value creation in cross-border acquisitions: the moderating role of cultural similarity. Journal of Management Studies 52(6): 766–795.

10.

Bergstrand

(1985) The gravity equation in international trade: some microeconomic foundations and empirical evidence. The Review of Economics and Statistics 67(3): 474–481.

11.

Bergstrand

Larch

Yotov

(2015) Economic integration agreements, border effects, and distance elasticities in the gravity equation. European Economic Review 78: 307–327.

12.

Björkman

Stahl

Vaara

(2007) Cultural differences and capability transfer in cross-border acquisitions: the mediating roles of capability complementarity, absorptive capacity, and social integration. Journal of International Business Studies 38: 658–672.

13.

Carril-Caccia

Paniagua

Requena

(2021) Asylum migration, borders, and terrorism in a structural gravity model. Politics and Governance 9(4): 146–158.

14.

Coeurdacier

De Santis

Aviat

(2009) Cross-border mergers and acquisitions and European integration. Economic Policy 24(57): 56–106.

15.

Coeurdacier

Martin

(2009) The geography of asset trade and the euro: insiders and outsiders. Journal of the Japanese and International Economies 23(2): 90–113.

16.

Correia

Guimaraes

Zylkin

(2020) Fast Poisson estimation with high- dimensional fixed effects. STATA Journal 20(1): 95–111.

17.

Dauvergne

(2004) Sovereignty, migration and the rule of law in global times. The Modern Law Review 67(4): 588–615.

18.

Dogru

Sirakaya-Turk

Crouch

(2017) Remodeling international tourism demand: old theory and new evidence. Tourism Management 60: 47–55.

19.

Egger

Tarlea

(2015) Multi-way clustering estimation of standard errors in gravity models. Economics Letters 134: 144–147.

20.

Eilat

Einav

(2004) Determinants of international tourism: a three-dimensional panel data analysis. Applied Economics 36(12): 1315–1327.

21.

Eriksen

(2016) Overheating. An Anthropology of Accelerated Change. London, UK: Pluto Press.

22.

Fally

(2015) Structural gravity and fixed effects. Journal of International Economics 97(1): 76–85.

23.

Franco-Bedoya

Frohm

(2021) Reduced ‘border effects’, free trade agreements and international trade. The World Economy 45(4): 1112–1139.

24.

Gil-Pareja

Llorca-Vivero

Martínez-Serrano , et al. (2007) The effect of EMU on tourism. Review of International Economics 15: 302–312.

25.

Glick

Rose

(2016) Currency unions and trade: a post-EMU reassessment. European Economic Review 87: 78–91.

26.

Hall

(2005) Tourism: Rethinking the Social Science of Mobility. London: Pearson Educational.

27.

Hageman

Berger

Gemie

, et al. (2004) Creating and Crossing Borders: The State, Future and Quality of Border Studies. Wales: Glamorgan: University of Glamorgan.

28.

Harb

Bassil

(2020) Gravity analysis of tourism flows and the ‘multilateral resistance to tourism. Current Issues in Tourism 23(6): 666–678.

29.

Head

Mayer

Ries

(2010) The erosion of colonial trade linkages after independence. Journal of International Economics 81(1): 1–14.

30.

Head

Mayer

(2014) Gravity equations: workhorse, toolkit, and cookbook. In: Handbook of International Economics: Elsevier, 4, pp. 131–195.

31.

Head

Mayer

(2021) The United States of Europe: a gravity model evaluation of the four freedoms. The Journal of Economic Perspectives 35(2): 23–48.

32.

Head

Ries

(1998) Immigration and trade creation: econometric evidence from Canada. Canadian Journal of Economics 31(1): 47–62.

33.

Keum

(2010) Tourism flows and trade theory: a panel data analysis with the gravity model. The Annals of Regional Science 44: 541–557.

34.

Khadaroo

Seetanah

(2008) The role of transport infrastructure in international tourism development: a gravity model approach. Tourism Management 29(5): 831–840.

35.

Khalid

Okafor

Burzynska

(2021) Do regional trade agreements enhance international tourism flows? Evidence from a cross-country analysis. Journal of Travel Research 61(6): 1391–1408.

36.

Liu

Hull

Luo

, et al. (2015) Multiple telecouplings and their complex interrelationships. Ecology and Society 20(3): 44.

37.

Llorca-Vivero

(2008) Terrorism and international tourism: new evidence. Defence and Peace Economics 19(2): 169–188.

38.

Maddux

Galinsky

(2009) Cultural borders and mental barriers: the relationship between living abroad and creativity. Journal of Personality and Social Psychology 96(5): 1047–1061.

39.

Martín

Prados

Jiménez

, et al. (2020) Interferences generated on the well-being of local communities by the activity of online platforms for tourist accommodation. Journal of Sustainable Tourism. DOI: 10.1080/09669582.2020.1861455.

40.

Mayer

Méjean

Nefussi

(2010) The location of domestic and foreign production affiliates by French multinational firms. Journal of Urban Economics 68(2): 115–128.

41.

Mayer

Zbaraszewski

Pieńkowski

, et al. (2019) Cross-Border Tourism in Protected Areas. Potentials, Pitfalls and Perspectives. Springer.

42.

McCallum

(1995) National borders matter: Canada-U.S. Regional trade patterns. The American Economic Review 85(3): 615–623.

43.

Morley

Rossello

Santana-Gallego

(2014) Gravity models for tourism demand: theory and use. Annals of Tourism Research 48: 1–10.

44.

Paniagua

Peiró-Palomino

Santana-Gallego

(2022) Does happiness drive tourism decisions? Economic Modelling 111: 105824.

45.

Parsley

(2003) Performing the border: Australia’s judgment of “unauthorised arrivals” at the airport. Australian Feminist Law Journal 18(1): 55–75.

46.

Patuelli

Mussoni

Candela

(2013) The effects of world heritage sites on domestic tourism: a spatial interaction model for Italy. Journal of Geographical Systems 15(3): 369–402.

47.

Popli

Akbar

Kumar

, et al. (2016) Reconceptualizing cultural distance: the role of cultural experience reserve in cross-border acquisitions. Journal of World Business 51(3): 404–412.

48.

Richardson

(2013) Borders and mobilities: introduction to the special issue. Mobilities 8(1): 1–6.

49.

Rowen

(2016) The geopolitics of tourism: mobilities, territory, and protest in China, taiwan, and Hong Kong. Annals of the American Association of Geographers 106(2): 385–393.

50.

Rumford

(2006) Theorizing borders. European Journal of Social Theory 9(2): 155–169.

51.

Salinas

Serdeira

Martín

, et al. (2020) Determinants of tourism destination competitiveness in the countries most visited by international tourists: proposal of a synthetic index. Tourism Management Perspectives 33: 100582.

52.

Salter

(2006) The global visa regime and the political technologies of the international self: borders, bodies, biopolitics. Alternatives: Global, Local, Political 31: 167–189.

53.

Salter

(2008) When the exception becomes the rule: borders, sovereignty, and citizenship. Citizenship Studies 12(4): 365–380.

54.

Santana-Gallego

Paniagua

(2022) Tourism and migration: identifying the channels with gravity models. Tourism Economics 28(2): 394–417.

55.

Santana-Gallego

Ledesma-Rodríguez

Pérez-Rodríguez

(2016) International trade and tourism flows: an extension of the gravity model. Economic Modelling 52: 1026–1033.

56.

Santos-Silva

Tenreyro

(2006) The log of gravity. The Review of Economics and Statistics 88(4): 641–658.

57.

Smith

SLJ

(1984) A method for estimating the distance equivalence of international boundaries. Journal of Travel Research 22(3): 37–39.

58.

Sofield

THB

(2006) Border tourism and border communities: an overview. Tourism Geographies 8(2): 102–121.

59.

Song

Cao

(2017) Tourism and economic globalization: an emerging research agenda. Journal of Travel Research 57(8): 999–1011.

60.

Timothy

(1999) Cross-border partnership in tourism resource management: international parks along the US–Canada border. Journal of Sustainable Tourism 7(3&4): 182–205.

61.

Timothy

(2001) Tourism and Political Boundaries. London & New York: Routledge.

62.

Timothy

(2005) Tourism and political boundaries in an age of globalization. In: Paper Presented at the Conference on Border Tourism and Community Development Conference. Xishuangbanna, China. 6–9 July.

63.

Timothy

Tosun

(2003) Tourists’ perceptions of the Canada–USA border as a barrier to tourism at the International Peace Garden. Tourism Management 24: 411–421.

64.

Tong

(2005) Ethnic networks in FDI and the impact of institutional development. Review of Development Economics 9(4): 563–580.

65.

Tubadji

Nijkamp

(2018) Revisiting the Balassa–Samuelson effect: international tourism and cultural proximity. Tourism Economics 24(8): 915–944.

66.

Umber

Grote

Frey

(2014) Same as it ever was? Europe’s national borders and the market for corporate control. Journal of International Money and Finance 40: 109–127.

67.

UNWTO – World Tourism Organisation (1991) Resolutions of the International Conference on Travel and Tourism. OttawaMadrid: CanadaUNWTO.

68.

UNWTO (2019) Tourism Barometer. Madrid: UNWTO.

69.

Van der Zee

Vanneste

(2015) Tourism networks unravelled; A review of the literature on networks in tourism management studies. Tourism Management Perspectives 15: 46–56.

70.

Von Bergner

Lohmann

(2014) Future challenges for global tourism: a delphi survey. Journal of Travel Research 53(4): 420–432.

71.

Weidenfeld

(2013) Tourism and cross border regional innovation systems. Annals of Tourism Research 42: 191–213.

72.

Więckowski

(2010) Tourism development in Polish borderlands. Geographia Polonica 83(1): 67–81.

73.

Więckowski

Timothy

(2021) Tourism and an evolving international boundary: bordering, debordering and rebordering on Usedom Island, Poland-Germany. Journal of Destination Marketing & Management 22: 100647.

74.

Wonders

(2006) Global flows, semi-permeable borders and new channels of inequality: border crossers and border performativity. In: Pickering

Weber

(eds), Borders, Mobility and Technologies of Control. Amsterdam: Springer, pp. 63–86.

75.

WTTC (2020) Travel & Tourism’s direct, indirect and induced impact. Available in: https://wttc.org/Research/Economic-Impact

76.

Wang

, et al. (2019) Modelling international tourism flows to China: a panel data analysis with the gravity model. Tourism Economics 25(7): 1047–1069.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.11 MB