Democracy and tourism demand in European countries: does environmental performance matter?

Abstract

Recently, empirical studies revealed that democracy is positively associated with environmental quality through the freedom gained by the people to demand environmental protection. In this paper, we explore empirical evidence linking how environmental performance interacts with democracy to influence tourism demand in twenty-seven European countries. To achieve this objective, we use the method of moments quantile regression (MMQR) model by Machado and Silva (J Econ 213: 145-173, 2019) and a balanced panel data covering the period 2002 to 2014. The empirical results suggest that environmental performance interacts heterogenously with democracy at different quantiles of the conditional distribution to stimulate tourism demand. Also, the effect of an increase in income and environmental performance is stronger in countries with lower tourism market shares than in countries with higher tourism market shares. The major implication for this study is that countries with lesser shares of the tourism market should strive for higher environmental performance and economic development as this would grant them more advantage in the tourism sector than their counterparts with higher market shares.

Introduction

The world has come to realize that to build a sound economy, much attention needs to be given to the development of the tourism industry, which enhances economic growth and job creation. Consequently, tourism development becomes a great concern of the government and its managers in most countries of the world. According to the recent statistics by the United Nations World Trade Organization (2019), the world total earning from international tourism increased over the years with an unprecedented growth of 6% in 2018, which is higher than the growth of the merchandize exports. In European countries, the travel and tourism sector alone generated about 14 million direct jobs in 2017 and 14.4 million in 2018. More so, the contribution of this sector to gross domestic product (GDP) of the continent significantly increased from 1843.1 USD in 2012 to 2155.5 billion USD in 2018. As tourism sector is developing, environmental quality may tend to deteriorate due to the effects of an increase in carbon dioxide emissions from excessive use of fuel oil and other traditional energy consumption patterns related to tourism development (See Katircioğlu 2014; Usman et al. 2019a).

Even though several studies have empirically examined the tourism-environment relationship (Dogan et al. 2017; Dogan and Aslan 2017) and tourism-economic growth relationship (Gunduz and Hatemi-J 2005; Balcilar et al. 2014; Shahbaz et al. 2018; Balcilar et al. 2020), the literature is still devoid of studies focusing the extent to which an increase in environmental quality would attract the inflow of tourism. In addition, these relationships may not be formed in isolation from the political institutions that govern the process. Hence, institutional variable such as democracy can affect tourism demand as revealed by Antonakakis et al. (2016). This argument is based on the positive correlation between democracy and income, which is the cornerstone of the modernization theory.¹ According to this theory, democracy may affect tourism sector in two ways: first, democracy increases income, which help in stimulating tourism development, and second, as income increases, environmental quality may deteriorate due to increase in energy consumption. On empirical ground, studies like Farzin and Bond (2006); Lv (2017); Usman et al. (2019b); Usman et al. (2020), and Ike et al. (2020) have provided evidence that the effect of democracy on macroeconomic variables changes with a country’s income level. According to Winslow (2005) and Mak Arvin and Lew (2011), in a democratic setting, environmental quality tends to improve because people are better informed about environmental issues. Consequently, the total number of international tourism arrival increases. On the other hand, studies like Heilbronner (1974), Midlarsky (1998), and Roberts and Parks (2007) argue that the income effect of democratic regime increases carbon dioxide emissions, which in turn, discourages international tourism arrival.

Given this background, the objective of this study is to examine how environmental performance interacts with democracy to induce tourism demand in twenty-seven EU countries² and determine whether this effect is consistent across countries with different tourism market shares. Our choice of using European data is predicated on the fact that the travel and tourism sector in Europe grows significantly over the years, with the sector generating millions of direct and indirect jobs as well as contributing heavily to their Gross Domestic Product (GDP).³ Furthermore, this period of tourism development coincides with the great improvement in democracy and accountability of the region. As such, our study makes a four-fold contribution to the literature. First, the paper reveals how income and the interaction of democracy and environmental performance enhance the government’s goal of developing the tourism sector. Second, our paper allows to ascertain if the effect of these variables would change as the market share of tourism increases across countries. Third, we apply a model that controls for both distributional and cross-country unobserved heterogeneity by incorporating fixed effects. Fourth, by employing the method of moments quantile regression (MMQR) recently proposed by Machado and Silva (2019), we provide insights into the distributional heterogeneity of the environmental performance-tourism demand nexus at different conditional quantile distributions of tourism demand. Unlike other panel quantile regressions developed by Koenker (2004), Lamarche (2010), and Canay (2011), the MMQR invariably assumes that the covariate only affects the distribution of the variable of interest via location and scale functions rather than just shifting locations. Hence, it allows individual effects to influence the entire distribution.

The rest of the paper is structured as follows: the “Data and methodology” section presents the methodology employed. the “Results and discussion” section analyze the result while the “Conclusion” section concludes the paper.

Data and methodology

This study employs a panel data from 2002 to 2014 for 27 European countries as shown in Appendix 1. The study period is selected based on data availability. The variables, measurements, and sources are shown in Table 1.

Table 1 Variable, measurement, and source

Modelling techniques

In this study, the MMQR with fixed effects recently proposed by Machado and Silva (2019) is applied. One of the main advantages of this method is that, it enables the researchers to capture the distributional heterogeneity of the environmental performance-tourism demand nexus at different conditional quantile distributions of tourism demand by incorporating fixed effect—an effect, which is unavailable in conventional mean regressions. Following Martins et al. (2017), we construct a standard tourism demand model as⁴:

$$ {QlnTD}_{it}\left(\tau |{X}_{it}\right)={\alpha}_0+ lnR{P}_{it}+{\alpha}_1\mathit{\ln} GD{P}_{it}+{\alpha}_2 DM{C}_{it}+{\alpha}_3 EP{I}_{it}+\left( DM{C}_{it}\ast EP{I}_{it}\right)+{\varepsilon}_{it} $$

(1)

From Eq. (1) lnTD_it(τ| X_it) represents τ^th conditional quantile function, and TD which is the dependent variable measures total international tourism receipts, ln is the natural logarithm, X_it denotes the explanatory variables. RP represents relative prices,⁵GDP measures the level of world income per capita; EPI measures environmental performance, DMC measures democracy, DMC ∗ EPI represents the interaction term of democracy and environmental performance while ε_it denotes the residual which is independently and identically distributed across individual country i at time t. A strong democracy can incentivize the flow of tourists due to perceived assurances bordering on safety as well as reduced human rights violations. Environmental quality can also spur tourist inflows because humans have a natural affinity towards healthier environments. Also, tourism arrivals can negatively affect environmental performance due to the possible environmental degrading effect of tourism activities. The residual is orthogonal to X_it and normalized in order to satisfy the moment conditions in Machado and Silva (2019) which do not imply strict exogeneity. Therefore, from Eq. (1), it implies that:

$$ \mathit{\ln} TD\left(\tau |{X}_{it}\right)=\left({\alpha}_i+{\theta}_iq\left(\tau \right)\right)+X{\prime}_{it}\beta +Z{\prime}_{it}\gamma q\left(\tau \right) $$

(2)

where α_i(τ) ≡ α_i + θ_iq(τ) is the scalar parameter which is indicative of the quantile-τ fixed effect for individual i. Z is a k-vector of identified components of X which are differentiable transformations with element l given by Z_l = Z_l(X), l = 1, …, k. Unlike the least squares fixed effects, the individual effects in this method do not represent intercept shifts. They are time-invariant parameters whose heterogeneous impacts are allowed to vary across the quantiles of the conditional distribution of the dependent variable. Equation (1), which is the conditional quantile function of the tourism demand-environmental performance nexus, is estimated using the MMQR approach, which gives solution to the following optimization problem:

$$ {\mathit{\min}}_q\sum \limits_i\sum \limits_t{\rho}_{\tau}\left({\hat{R}}_{it}-\left({\hat{\delta}}_i+{Z}_{\mathrm{i}t}\overset{\acute{\mkern6mu}}{\gamma}\right)q\right) $$

(3)

where ρ_τ(A) = (τ − 1)AI{A ≤ 0} + τAI{A > O} is the standard quantile loss function. Due to a marginal change in i, the parameter for the dependent variable i may signify the marginal change in the r^th conditional quantile of lnTD_it(τ| X_it).

Furthermore, we employ alternative estimation techniques to ascertain if the parameter estimates are robust to cross-sectional dependence as described by Acaravci and Akalin (2017). Due to the possible distorting effect of cross-sectional dependence and auto-correlation, we employ the fixed effects OLS (FE-OLS) and the random effects GLS (RE-GLS) regression with Driscoll and Kraay (1998) standard errors, which are robust to general forms of cross-sectional dependence and auto-correlation up to a specified lag. If the parameters from the RE-GLS and the FE-OLS mean regressions correspond closely to the location parameters of the MMQR in terms of magnitude and significance, it then implies that the MMQR estimation is robust to cross-sectional dependence and auto-correlation (see Machado and Silva 2019).

Results and discussion

The results from Table 2 indicate that the state of the world economy affects tourism demand positively and significantly, across all quantiles; however, the scale of this effect reduces from the lowest to the highest quantiles. This is also confirmed from the scale parameters. The elastic world income effect suggests the perception by the world that tourism to EU countries is a luxury good. An increase in relative prices diminishes tourism demand across all quantiles. Environmental performance and democracy are also shown to have positive but nonlinear effects on tourism. An increase in democracy increases tourism demand in countries with less environmental performance but reduces it in countries with higher environmental performance. Also, an increase in environmental performance increases tourism demand in countries with weaker democracy but reduces tourism demand in countries with stronger democracy. This result is consistent with Antonakakis et al. (2016) which isolates an economic driven tourism demand relationship in nondemocratic countries. This relationship may allude to the possibility that in countries with either higher environmental quality or stronger democracy or both, the market for tourism may be more saturated than in countries in which these variables are relatively weaker. The effect of both environmental performance and democracy are, however, both insignificant at the highest quantiles (8th to 9th) which may entail the saturation of the tourism market in countries with the highest tourism market shares. The result aligns with Usman et al. (2019b) who found support for the hypothesis that democracy provides freedoms, which may improve environmental quality through a positive income effect. Also, the finding is similar to Neumayer (2004) who reported that tourism demand in autocratic regimes is low due to human rights violations, terrorism, and conflict intensity and Saha and Yap (2014) who discovered that political instability associated with autocracy is a barrier to tourism demand and thus strengthening democracy in these countries would see a surge in tourism demand. The findings are validated by the robustness checks via (FE-OLS) and the GLS (RE-GLS) regressions with Driscoll-Kraay standard errors.⁶

Table 2 MM-QR estimation results

Conclusion

The objective of this paper is to investigate how democracy interacts with environmental performance to induce tourism demand in EU-27 countries by employing the novel MMQR approach proposed by Machado and Silva (2019). The results suggest that democracy spurs tourism in countries with lesser environmental performance while environmental performance spurs tourism in countries with weaker democracy alluding to the possibility of tourism market saturation in countries where either of these variables is at high levels. The quantile estimates show that countries with lower tourism market shares are more sensitive to the state of the world economy than countries with lesser tourism market shares. The policy implication for our findings is that countries with lesser tourism market shares should develop the quality of their environment and strengthen their democracy as this is the surest way to significantly improve the tourism sector.

Appendices

Appendix 1

Table 3 List of countries

Appendix 2

Table 4 Metrics for measuring a country environmental performance