The role of international tourism on foreign trade in the BRICS nations

Abstract Does international tourism improve foreign trade in BRICS countries? A handful of empirical literature has attempted to answer this profound research question. However, the results of these studies are inconclusive. Thus, the aim of the study was to contribute to the scant empirical literature on international tourism and foreign trade by examining the impact of tourism on foreign trade in BRICS countries. To achieve this aim, the study employed the panel autoregressive distributive model (P-ARDL). The P-ARDL results reveal that tourists arrival, receipts from tourism exports and economic growth are significant and positively influence foreign trade. Noteworthy is that international tourism plays a significant role in the foreign trade. Thus, further assimilation and orchestrating of trade and tourism policies inside the BRICS members should be promoted.


Introduction
The tourism sector has occupied a large economic space worldwide. This notion is illustrated by an increase in the number of tourists from 20 million to 1.5 billion in 1950 and 2019, respectively (World Tourism Organisation, 2019b). Tourism proponents claim that an increase in tourism activities is a response to the globalisation call (Zhuang et al., 2019). Globalisation has ushered the free movement of goods and services which has also made foreign trade simpler. Thus, emerging countries have shifted their prerogative from traditional sectors to modern sectors such as trade and tourism due to high productivity in the sectors and globalisation. These ABOUT THE AUTHOR Rufaro Garidzirai is a structured academic professional with eight years' experience in the academia. My research work is premised on three economics fundamental themes namely: development economics, macroeconomic policy, and international economics. My zeal in these themes is specifically directed on economic growth, local economic development, inequality, trade, and poverty. My philosophy has been to find the answers to these socio-economic challenges that affect developing and emerging countries such as South Africa. Over the past eight years my publications have sought to address these socio-economic challenges. These publications comprise of twenty-five peer reviewed articles, two book chapters, and two conference publications.

PUBLIC INTEREST STATEMENT
International tourism is a heuristic cue for international trade. Economic theory supports this relationship and implies that the relationship has the potential of improving other key important sectors in an economy. Since the empirical literature is scant, the study covered this research lacuna using a Pooled Mean Group. The results of the study imply that tourism is an important vehicle for trade. Noteworthy is that international tourism plays a significant role in the foreign trade. Thus, further assimilation and orchestrating of trade and tourism policies inside the BRICS members should be promoted. economic sectors are attached to positive externalities such as an increase in tourism receipts, increased competition, improved economic growth, creation of jobs and poverty reduction (Zhao and Xia, 2019;Saayman et al., 2012).
Given the substantial monetary and fiscal contribution of the tourism sector, it has been identified as one of the sectors that can improve the Sustainable Development Goal number 17. The seventeenth goal seeks to achieve international trade worldwide, that is to increase the number of exports (Sustainable Development Goals, 2020). The researcher opines that tourism is a vehicle to achieve this objective. This notion is in line with the diffusion theory. In this light, Batta (2000) claims that international tourists buy local goods and services, which leads to an increase in the exports in the local country and imports in the destination country. Thus, tourists buy the host country's exports which bring income to the host country and eventually improves economic growth through the multiplier process (Kačar & Ikić S, 2016;Mirică (Dumitrescu), 2019). This is achieved through the purchasing habits by consumers. It has come a norm that tourists have a spending habits that make them purchase souvenir etc . Empirical literature also confirmed the diffusion theory (Chaisumpunsakul & Pholphirul, 2018;Faber & Gaubert, 2019). These studies concluded that international tourism increases international trade by at least 1%.
Considering that the BRICS is one of the strongest trade partnership worldwide, it has attracted the attention of researchers especially on the relationship between tourism and international trade (Chaisumpunsakul & Pholphirul, 2018;Shakouri et al., 2017;Viljoen et al., 2019). Interestingly, these authors have bewailed the scant academic literature on trade and tourism. The limited existing literature is believed to be erratic and inconclusive considering the importance of the relationship between tourism and trade (Faber & Gaubert, 2019;Guo & Liao, 2017). Since, the link between foreign trade and tourism is scarce and inconclusive, the current study addresses this research gap by examining the relationship between foreign trade and tourism in the BRICS nations with the purpose of contributing to the body of literature on global trade and tourism. The BRICS nations were chosen as the study area because of the scarcity of tourism research in the area, regardless of the fact that it is one of the influential trade partnerships worldwide. Furthermore, the BRICS nations were chosen since their partnership is a trade agreement and the study is based on international trade. According to the knowledge of the author, there is no study that has investigated such a relationship on the BRICS bloc. Therefore, the objective of this study is to examine the impact of international tourism on international trade. This study is set to make three contributions. The first contribution is that the variables under study have not been examined in the BRICS countries. This study to the researcher's knowledge is the first study that analyses the impact of tourism on international trade, thus contributing to the scant literature on tourism and trade body of literature. Secondly, the researcher modelled international trade as function of international tourism unlike majority of the studies that used other macroeconomic variables such as economic growth, poverty, and inequality. Another notable contribution is that the paper posits a new cointegration model namely the Panel Autoregressive Distributive model-P-ARDL (Pesaran et al., 2001). There is overwhelming evidence in the literature that stipulates that P-ARDL model is the new cointegration model (Dube & Zhou, 2013;Haug, 2008;Mokotsanyane, 2016;Nosier, 2012;Samargandi et al.).
The rest of the paper is arranged in the following manner. Section 2 gives the BRICS stylized facts on the variables under study. Section 3 and 4 present the study's methodology and results. Conclusion was presented on section 5.

Literature review
The association between international trade and international tourism can be linked with the diffusion theory and gravity theory. The diffusion theory positions the tourism sector at the center of international trade. The theory follows the analogy that international tourists buy local goods and services which increases exports in the local country and imports in the destination country (Batta, 2000). This means that international tourists that arrive in a host country spends on local goods and services that also tends to increase the international receipts of the host country. The theory exponents argue that the link between exports and tourism has a multiplier effect that influences the trade balance. Thus, if the exports are more then the trade balance will eventually improve for the better. This has an opposite effect when international tourists leave the country. Exports will decrease significantly. Moreover, the diffusion theorists opine that the tourism sector is developmental in nature. Thus, other economic sectors such as the agriculture and transport sector benefit from the tourism sector. Noteworthy is that the gravity model further endorses the diffusion theory. It stipulates that the link between tourism and international trade is more effective when the countries under consideration have something in common (Saayman et al., 2012). For instance, the BRICS bloc has common features such as: trading partners, emerging economies, large countries to mention but a few. The theorists argued that it is easy to visit such countries and conduct business in such countries. For instance, it becomes easier for the BRICS countries to trade without trade restrictions. Such trade conditions granger cause tourism.
Of paramount importance is that this study cannot be separated from other previous studies such as (Chaisumpunsakul & Pholphirul, 2018;Faber & Gaubert, 2019: Guo & Liao, 2017Shakouri et al., 2017). For instance, Chaisumpunsakul & Pholphirul, 2018) examined the association between trade openness and Thailand international trade partners. The authors used trade openness as a dependent variable while tourism arrival was independent variables. The study utilized the dynamic Panel Data Estimator or GMM-SYS estimator and found that international trade positively influences international tourism. The study found that a 1% increase in international trade increases international tourism by 0.6%. Faber and Gaubert (2019) investigated the relationship between tourism and international trade in Mexico. The study used the microdata and a quantitative spatial equilibrium model and found that tourism receipts and arrival improve economic growth and international trade. The authors further found that the tourism sector has spillover effects on the manufacturing sector. Contrary results were found by Habibi and Ahmadzadeh (2015) who examined the association among economic growth, trade, and tourism in Malaysia from 1980 to 2013. The study employed quarterly data and the ARDL model and the results show that there is no significant relationship between tourism and international trade in Malaysia. However, a positive relationship between tourism and economic growth was observed. One can observe that the studies have come up with different results. This has been attributed on the fact that both studies have used a different methodology and different proxies. Furthermore, the results were also influenced by tourism and trade policies used in each country. Shakouri et al. (2017) examined the link among tourism, economic growth, and foreign trade in the selected Asian countries from 1995 to 2014. The authors measured tourism using international tourism receipts and tourism arrivals. The main purpose of the study was to investigate the relationship among the relationship using the variance decomposition and panel Granger causality analysis. The results of the study reveal that tourist's arrival and receipts positively influence economic growth and foreign trade. The authors argued that an increase in the number of tourists increases the purchase of local goods by international tourists. Using, the Granger causality test, Ozcan (2016) examined the causal association between tourism and international trade between 1995 and 2013. The author used tourism arrival as a proxy of tourism while foreign trade was used a proxy of international trade. The study focused on selected Mediterranean countries and found that tourism granger cause trade. Conversely, in a different set-up of G20, an inverse relationship between international tourism and foreign trade was found by Lu. The authors measured tourism using tourists' arrival and receipts under panel data and found that these have a direct on foreign trade. Santana-Gallego et al. (2011) also used the panel data technique to examine the relationship between international tourism and foreign trade in the OECD countries. The study found that inbound tourism promotes foreign trade. The author further reiterated that the departure and arrival of international tourists leads to the smooth flow of goods and services. Guo and Liao (2017) further highlight that there are other factors that influence trade and international tourism. The authors identified the consumption rate, economic growth, foreign exchange, and government macroeconomic regulations as factors that influence tourism and international trade. The authors argued that a stable exchange rate attracts international tourists and contributes significantly to the trade balance. Furthermore, a country with a better economic growth attracts international tourists and granger causes foreign trade. These factors were further confirmed by the study done by Culiuc (2014) who investigated the impact of macroeconomic factors on tourism and foreign trade in the OECD. The study utilized the gravity model and found that economic growth and exchange rate play a significant role in the tourism and foreign trade space. Raspor et al., 2017 analysed the tourism revenue and exports in China, Slovenia, and Montenegro. The study employed the methods of analysis and comparative analysis and found a positive relationship between China's exports and tourists from Montenegro and Slovenia. The authors argued that consumption expenditure increases and improves trade balance. A similar study was conducted in Europe by Castro (2016) who examined the link among trade, economic growth, and the tourism sector in Spain. The study found an inverse relationship between international trade and tourism, implying that as the number of tourists increases, the number of goods and services in a country decrease. Different results were found by 2012) also used the Granger Causality test to find the causal link between international trade and tourism. The study concluded that the relationship between trade and tourism is significant and bidirectional. Thus, tourism causes the flow of goods from a host country to a destination country, thereby improving the trade balance and economic growth of the respective countries. In Africa, a study by Viljoen et al. (2019) was conducted. The authors investigated whether the trade theory explains Africa's tourism in 25 countries. To achieve this aim, the study employed a panel data analysis and found that tourism receipts have promoted international trade in the African countries since these countries have a comparative advantage in tourism and international trade globally. Another study that used a gravity model found that population is positively related to foreign trade. This study was based in Pakistan from the period 2006 to 2015. These results are also in line with the study done by Peterson (2017) that utilized historical data charts.
The empirical work reviewed reveals that majority of the studies focused more on the influence of international trade on tourism. The focus was to assess whether international trade has the ability of bringing tourists in a host country. Without any doubt, most of the studies point to a positive impact of international trade on tourism arrival. Relatively, there is a notably few academic inquiries into the impact of tourism on foreign trade. The few studies done seem to be infrequent and lacks sufficient depth, thus, the study seeks to address this discernable gap by employing a unique Panel Autoregressive Distributive Lag Model (P-ARDL) under Pooled Mean Group. This methodology gives the researcher advantage because the dataset can be pooled and averaged and allows the short-run coefficients to differ across countries. According to the knowledge of the author, there are no scholars that have investigated this relationship and used P-ARDL in the BRICS bloc.

Stylized facts on BRICS
The BRICS is a trade agreement that consists of five countries, namely, Brazil, Russia, India, China, and South Africa (International Trade Centre, 2017). This trade agreement was founded in 2006 and by then it was called BRIC: Brazil, Russia, India, and China (Neethling, 2017). South Africa later joined the association in 2010. Measey et al. () mention that the BRICS bloc is composed of emerging economies that have fast growing economic sectors. For the purposes of this study, the researcher focused on the BRICS economic growth, international trade, and tourism. For instance, the BRICS have contributed about 20% of global GDP in 2019 and 50% of the global GDP in the past decade. In other words, the BRICS bloc dominates the global economic growth. The bloc has a comparative advantage in trade, manufacturing, ICT sector, and primary commodities.
These sectors are expected to contribute more to the GDP bloc considering the introduction of the fourth industrial revolution.
In terms of trade, Rasoulinezhad and Jabalameli (2018) mention that the BRICS nations have made a significant impact to global trade. The bloc has tripled the trade statistics over the last two decades (National Treasury, 2020). For instance, the exports value has been increasing since the formation of this trade bloc. This is shown by an increase in exports value from 8% in 2006 to 18 percent in 2019 (World Trade Organization, 2020). On the other hand, the BRICS is contributing about 15% of imports to the global imports. This means that the BRICS trade is greatly synchronized. For example, South Africa, Russia, and Brazil are rich in natural resources and commodity prices. India and China export these goods from South Africa, Russia, and Brazil, while India has a comparative advantage in analgesics products, information communication technology, and fabrics. China's strength, on the other hand, lies in the manufacturing sector (World Bank, 2019). It is important to note that China is the largest exporter of manufactured goods and services globally (World Trade Organization (2019). Industrial Development Corporation (2018) and Maryam et al. (2018) mention that the BRICS excel in trade due to an abundance of natural resources, cheap labour, and specialization. Thus, the bloc has been experiencing a sustained positive trade balance compared to the rest of the world.
Having considered the BRICS and the trade sector, it is prudent for one to consider the BRICS and the tourism sector as well. Worth noting is that, the BRICS bloc has a comparative advantage in the tourism sector since the number of tourist arrival keeps on increasing in the respective BRICS countries. The World Bank, 2019) reports that an increase in the BRICS tourists was ascribed to the 2010 soccer world cup in South Africa, 2014 soccer world cup in Brazil, and 2018 Russia World cup . Furthermore, India has also hosted many cricket tournaments together with China hosting many Olympics games. PwC (2020) reports that the number of tourists in the BRICS has increased by an average of 5%, yearly, due to the less stringent travel conditions in these countries. Moreover, tourists visit the bloc due to the beautiful scenery in these countries since all the countries are in the top 20 tourist attraction countries (World Tourism Organisation, 2019b). The increase in the number of tourist arrival has also improved the BRICS international trade. Hence, if international trade is well harnessed in these countries, it has the potential of improving economic development in each country.

Research data and specification
To investigate the association between international tourism and foreign trade in the BRICS nations, the study sourced data from World Bank (1995Bank ( -2017. The range of the study's period  was preferred due to the availability of the variables data. The data used in this study includes foreign trade (dependent variable), international tourism arrival, international tourism departure, international tourism receipts from exports and economic growth. Noteworthy is that economic growth was used as a control variable. To illustrate the above mentioned relationship, equation 1 was formulated.
Where lntrade i;t represents foreign trade in the BRICS countries, lntarrivals i;t is the number of international tourists in the BRICS countries, lntpop i;t is the number of people in BRICS country, lnxreceipts i;t is the exports receipts from international tourists, lngdp i;t is the economic growth in the BRICS countries, i represents each country in the BRICS countries, while t is the time and ε i;t is the error term.
Five variables were used in this study, namely: foreign trade, international exports tourism receipts, international arrival, international departure, and economic growth. Table 1 shows all the variables used in the study. Noteworthy is that foreign trade was used as a dependent variable since it captures goods and services traded by countries (World Bank, 2020). This measure is inline with the diffusion theory that stipulates that tourism diffuse goods and services from one nation to the other. There are several researchers that have used this proxy. International exports tourism receipts are defined as all the international tourist consumption expressed as a percentage of exports (Shakouri et al., 2017). The study expects a positive relationship between foreign trade and international tourism receipts from exports. The rationale is that the purchase of goods and services by tourists increases exports and thus improves the trade balance in the BRICS. International tourist arrival was also used as a dependent variable and defined as the number of tourists arriving in the BRICS countries (Tourism Statistics, 2020). The study expects international tourist arrival to positively influence foreign trade since the arrival of tourists increase the probability of buying local goods. In addition, international tourist departure is the number of tourists leaving the BRICS countries (Govdeli and Direkci; Van der Schyff et al., 2019). The author expects an inverse relationship between international departure and international trade. The rationale is that international tourist departure reduces the exports in a country that tend to affect foreign trade negatively Economic growth is an improvement in the market value of goods and services produced in the BRICS countries from one period to another (Feldman and Storper, 2011). This was measured by Gross Domestic Product per capita (Sekwaila & Garidzirai, 2020).

Estimation techniques
Prior to the actual research method, the study conducted the descriptive statistics and the stationarity test. The descriptive statistics was used to pronounce the features of the variables under study. The features are best explained using maximum, minimum, mean, kurtosis, and standard deviation. Secondly, the study conducted the panel unit root test. Like the time series unit root test, the purpose of the test is to determine the order of integration and methodology of the study. Hurlin and Mignony (2005) suggested three most important panel unit root tests, namely, the Levin et al. (2002), lm, Pesaran (2003, ADF of Maddala and Wu (1999). All these three tests set a null hypothesis of non-stationarity. Thus, the probability value of less than 10% rejects the null hypothesis and concludes that the variable is stationary.
Of note is that the unit root test gives a clear indication of the estimation technique to employ. Mohr and Fourie mention that a panel least square methodology is employed when all the variables under study are stationary at levels. The author further outlines that a Panel ARDL model is appropriate if the variables under study are integrated at level zero and one. A panel VAR or VECM methodology is used when the variables under study are integrated at level one. For this study, a panel ARDL model was employed since the variables were a combination of zero and one.
The study utilized the Panel-ARDL model recommended by Pesaran et al. (1999). The Panel-ARDL may also be called Pooled Mean Group. In this study, the words will be used interchangeably. This estimation technique was employed since the variables under the study are integrated at different levels. In addition, the P-ARDL was preferred since it is one of the latest panel estimation techniques which is also the new cointegration test (Dube & Zhou, 2013;Pesaran, 2004). Simply put, one does not need to conduct the cointegration tests. There are several studies that have used the same technique (Dube & Zhou, 2013;Haug, 2008;Mokotsanyane, 2016;Nosier, 2012;Samargandi et al.). The researchers observed that the P-ARDL model allows for an analysis of a short-run and long-run relationship to be conducted. A long-run analysis considers the sign of the coefficient and the probability value. This means that the probability value should be significant at one, five or 10%. On the other hand, a short-run relationship analyses the cointegration equation (error term) that should be negative and significant at 1%, 5% and 10%. The long-run and short-run equation is illustrated below: Where lntrade represents the trade balance in the BRICS countries, while X stands for all the independent variables. δ and γ represent the short-run coefficients of dependent and independent variables, respectively. The subscripts i and t stand for cross-section and time, respectively, β stands for long-run coefficients, while u stands for fixed effect and e is the error term.
The common problem of panel data is cross-sectional dependency. This problem causes spurious results. However, this can be avoided by testing the cross-dependency test using the Pesaran LM test, Breusch and Pagan and Baltagi. The probability of these tests should be below 10% for them to pass the cross-dependency test.

Empirical results
The empirical results report the descriptive statistics, unit root tests, P-ARDL, and post-estimation techniques. The following section discusses the descriptive statistics.

Stationarity test results
The panel unit root tests are reported in Table 2. The results reveal that lntarrival and lnxreceipts are stationary at levels. Therefore, we reject the null hypothesis since their probability values are less than 0.10. Other variables such as lngdp, lntrade, and lntdep were found not to be stationary at level, thus, there were first differenced. The panel unit root tests show the probability values of less than 10% confirming the stationarity at level one. As pointed out in section 3, a combination  of variables integrated at different levels, justifies the use of the P-ARDL model and there is no need to conduct cointegration tests (Pesaran, 2004).

Discussions of long-run analysis
This part is twofold: a long-run and short-run analysis. The following section discusses the long-run analysis estimated using Akaike info Criterion (1,1,1,1,1). Table 3 presents the long-run relationship foreign trade and international tourism estimated by three estimators, namely Pooled Mean Group (PMG), Mean Group (MG) and Dynamic Fixed Effects (DFE). To determine the best model to use, a Hausman test was employed, and the results illustrate Chi-Square p-value of 0.391 that is greater than 10%. Therefore, a null hypothesis cannot be rejected, and we can conclude that P-ARDL is the preferred model as it is more efficient and consistent compared to other estimators.
The PMG results reveal that lntarrival, lngdp, and lnxreceipts are significant and positively related to foreign trade. For example, a 1% increase in the lntarrival increases the trade inflow of goods and services by 0.87%. The study expected this result since the arrival of international tourists increases the probability of an increase in the consumption of local capital, goods, and services. This is in line with the diffusion theory that positions the tourism sector at the center of international trade.
A similar study by Chaisumpunsakul and Pholphirul (2018) and Shakouri et al. (2017) found such a relationship and claimed that international tourists are one of the major sources of trade flow in a macroeconomic model.
Likewise, a 1% increase in the income from international tourism receipts expressed as percentage of exports increases the trade flow of goods and services by 1.03%. This implies that the expenditure by international tourists constitutes a revenue to a host country in the form of an injection. This will increase the number of exports thereby improving trade and other sectors. This is in line with the diffusion theory that stipulates that tourism sectors have the spill-over effects to other economic sectors. These findings are in snyc with the studies done by, Faber and Gaubert (2019), Raspor et al., 2017) and Viljoen et al. (2019) found similar findings. The authors found that an increase in the purchase of exports in the BRICS countries improves the flow of goods and services.
Another prior expectation was met by the lngdp which is significant and positively influencing foreign trade. Thus, a 1% increase in the lngdpp increases the trade by 1.11%. This implies that the increase in goods and services is mainly influence exports and when the growth is also export based, a surge in the demand for imports will increase more than the aggregate demand that will cause exports to increase more than the increase in the imports. Lu and Shakouri et al. (2017) also endorsed such findings though the study was done in Asia. Interestingly, the lntpop was found to be significant and positively related to foreign trade. Thus, a 1% increase in population increases foreign trade by 0.5918. This implies that an increase in population increases the number of entrepreneurs, and consumers that increases the trade activities. The results are in sync with the studies done by Yeo and Deng () and Peterson (2017).

Short run analysis
Having established the long-run association between foreign trade and international tourism, the study analysed the short-run relationship. Table 2 shows the short-run results estimated using the PMG, MG, and DFE. Since, PMG was preferred the study discussed the short-run results using this estimator. The results in Table 4 reveal a significant and negative coefficient Error Correction Term (−0.2330). The negative error correction model confirms that the model converges in the upcoming years. Thus, all the macroeconomic shocks to be restored to equilibrium. Noteworthy is that tourism departure that was negatively related to foreign trade in the long-run is now showing a positive association. All other variables maintained a positive relationship both in the short-run and long-run. Interestingly, all the three estimators (PMG, MG, and DFE) results seem to show similar results.

Cross-sectional dependency test
To check for stability of the model employed in the study, a cross-dependency test was used and the results are illustrated in Table 5. The probability of Breusch-Pagan Chi-Square, Pearson LM, and Pearson CD are less than 10%. Thus, the study produced robust results and the model was stable.

Conclusion
The association between foreign trade and international tourism is an egg or chicken question in the tourism and trade debate. Thus, an increase in the number of tourists increase the flow of goods and services from one country to another. On the other hand, the flow of goods and services increase the number of tourists. Such a relationship is illustrated by the number of international tourists that visited the bloc during the World Cups and Olympics in 2010, 2014, 2016, and 2018. Notable is that international trade also increased significantly during these years. Therefore, such a relationship cannot be overlooked since tourism is a potential catalyst of international trade. As highlighted in section one, the objective of this study was to examine the impact of international  tourism on international trade. Hence, this paper studied the impact of international tourism on international trade in the BRICS countries. The preceding results from section 4 show international tourism as a pivotal factor that contributes to foreign trade. The researcher claims that an increase in the number of international tourists increases the consumption of goods and services, thereby, increasing the number of exports in a local country. This is in line with the empirical literature and the theoretical literature. Thus, the BRICS countries should take advantage of tourism, the fastest growing sector, in achieving the Sustainable Development Goal 17. Based on the study's results, a further assimilation and orchestrating of trade and tourism policies inside the BRICS members should be promoted. This includes giving the special investment inducements such as intermingled tourism and trade funds. Such a fund provides special skills and other technical skills to advance trade within BRICS. Although the study's objective has been accomplished, the researcher has identified a limitation. The study should have incorporated more countries and variables such as exchange rate. However, these limitations are to be addressed in future studies.

Funding
The author received no direct funding for this research.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Citation information
Cite this article as: The role of international tourism on foreign trade in the BRICS nations, Rufaro Garidzirai, Cogent Social Sciences (2022), 8: 2076792.