Economic impacts of tourism demand on the European food and accommodation industry

Abstract

Sustainability strategies have become central topics in the literature on tourism development, given the significant increase in tourism demand in Europe. This study examines the economic (cost) impact of increased tourism demand on the European Food and Accommodation Industry. We employed three distinct models — pooled linear regression, flexible nonparametric, and semiparametric smooth coefficients — using data spanning European countries from 2005 to 2020. Our findings reveal the heterogeneous impacts of tourist arrivals and length of stay on costs. Although the direct effects were predominantly positive, the indirect effects were found to be negative, resulting in a minor overall negative effect on the average. Furthermore, we found that the proportion of domestic tourists has a notable impact on costs, revealing that higher proportions are linked to lower direct costs but higher indirect costs on average. These findings provide valuable insights for industry stakeholders and policymakers alike.

Keywords

tourism demand costs domestic versus international tourism semiparametric smooth coefficient model

Introduction

The global economy has been significantly influenced by the tourism industry, drawing millions of visitors to diverse locations for distinctive travel experiences. Renowned European destinations like Spain, Italy, and Iceland have witnessed substantial tourist arrivals because of their abundant cultural heritage and picturesque landscapes (Streimikiene et al., 2021; Capocchi et al., 2019; Oklevik et al., 2019; Perkumienė and Pranskunienė, 2019; Seraphin and Ivanov, 2020; Séraphin et al., 2019; Wall, 2020). This surge in tourism demand has profound economic implications, especially for the food and accommodation sectors, which are crucial in providing lodging and dining experiences to tourists. Given the tourism sector’s growth and its importance to the European economy [see Table 1], understanding its impact on the food and accommodation industry is essential for policymakers, industry stakeholders, and researchers. Sustainability has emerged as a critical concern for tourism planners and policy makers (Kee and Chau, 2020; Pulido-Fernández et al., 2019). Achieving economic benefits while mitigating environmental and social consequences is paramount. This study explores the economic sustainability of tourism, focusing on the dynamic interaction between economic and environmental sustainability. Supporting this synergy, Zhu et al. (2021) establish a positive correlation between environmental sustainability and economic health, advocating a holistic approach that addresses all three dimensions simultaneously. However, short-term trade-offs between these dimensions require further exploration of economic sustainability (Kee and Chau, 2020). Balancing these aspects is essential to avoid negative impacts from unsustainable (over) tourism in Europe, a matter highlighted by bodies like the European Parliament (Peeters et al., 2021).

Table 1.

Descriptive statistics.

Variable	Labels	Mean	Min	1st Qu.	Median	3rd Qu.	Max.
Inputs
X1	Labor	1403.30	55.70	449.80	1181.50	2697.80	28209.90
X₂	Gross investment	603265.30	1214.50	89659.90	434050.70	2186207.70	92774641.50
X₃	Materials	2.99	1.66	2.31	3.01	3.85	5.29
Outputs
Y₁	Hotel and similar establishments	1,689.18	69.41	741.00	1,483.26	2,924.85	31,697.72
Y₂	Holiday and other short stay	161.16	1.40	47.90	113.20	484.54	5,584.22
Y₃	Camping recreation and similar	27.03	0.20	3.55	28.70	146.35	2,601.91
Y₄	Other accommodations	24.90	0.20	9.85	30.20	97.03	695.06
Y₅	Restaurant and mobile food services	2,901.55	154.47	769.70	3,029.04	6,815.81	52,052.08
Y₆	Event catering and other food services	465.45	9.30	86.49	527.42	1,245.14	14,115.09
Y₇	Beverage and serving	572.49	21.69	178.75	337.31	1,740.63	20,211.36
t	Year dummy
Tourism
Z₁	Total tourist arrival	13.94	0.87	5.90	12.24	24.89	185.77
Z₁₁	Domestic tourist arrivals	6.99	0.24	2.24	7.28	13.94	146.13
Z₁₂	Foreign tourist arrivals	5.72	0.51	2.35	4.64	12.48	67.66
Z₂	Total length of stay	2.82	1.58	2.35	2.69	3.41	5.83
Z₂₁	Length of stay domestic	2.58	1.56	2.34	2.54	2.95	4.66
Z₂₂	Length of stay foreign	2.93	1.61	2.23	2.63	3.83	6.66

The dataset comprises 248 observations in total. In the empirical analysis, logarithmic transformations were applied, except for the time variable t. Also, note that inputs, outputs and tourist arrivals statistics are provided in millions of Euro.

Recent research has shown significant interest in research on the economic impacts of tourism demand on the European food and accommodation industry. A surge in tourism demand is expected to yield both positive direct and indirect economic benefits in this sector. Studies indicate that as the number of tourists increases, there is increased demand and revenue for hotel accommodation, restaurants, cafés, and other food and beverages (Allan et al., 2022; Brida et al., 2015; Dritsakis, 2012; Gössling et al., 2020; Song et al., 2018). In addition, the increase in tourism demand stimulates employment opportunities and investments in infrastructure, enhancing economic activity across various sectors (Allan et al., 2022; Figini and Patuelli, 2022; Song and Li, 2008). However, disparities in tourism demand across European countries can lead to challenges such as overtourism, where popular destinations face negative environmental and economic impacts due to excessive tourists (Figini and Patuelli, 2022; Gössling et al., 2020).

Although tourism demand can enhance the economic prospects of the food and accommodation industry, it also has potential negative consequences. One of the challenges is the seasonal variation in tourist arrivals throughout the year. During peak seasons, overcrowding and pressure on resources are common, whereas, in off-peak periods, businesses in this sector face difficulties in maintaining profitability (Bramwell and Lane, 2000; Corluka, 2019). Furthermore, overtourism, caused by an excess of tourists beyond a destination’s capacity, incurs multiple economic costs, including disrupted local economies, reduced quality of life for residents, and environmental degradation. Studies such as Vizek et al. (2024) and Gössling et al. (2012) have highlighted negative impacts like the conversion of housing stock into tourist accommodations, higher housing prices, and increased water consumption.

Despite existing research, there are gaps in macro-level economic sustainability studies, often overlooking the economic dimensions of overtourism (Capocchi et al., 2019; Kee and Chau, 2020; Oklevik et al., 2019; Qiu et al., 2019). Economic sustainability research focuses primarily on GDP, employment, competitiveness, and environmental trade-offs (Garrigós-Simón et al., 2015; Kristjánsdóttir et al., 2018; Pulido-Fernández et al., 2015). The literature on unsustainable tourism predominantly addresses non-economic dimensions, yet the impact on economic sustainability remains crucial for businesses and regulators. Businesses must align their operations with the broader sustainability goal of balancing tourism supply and demand (Pulido-Fernández et al., 2019). Innovative strategies from destination managers and operators, as highlighted by Dwyer et al. (2009), can control customer behavior and promote environmentally responsible choices. However, limited attention has been paid to distinguishing between the impacts of domestic and foreign tourists. Only a few studies, including those by Larsen and Wolff (2019), Incera and Fernández (2015), and Muštra et al. (2023), have examined tourist satisfaction, income inequality, and the role of cultural heritage in economic resilience. However, the impact on economic sustainability from a cost perspective remains largely unexplored.

Therefore, this study investigates the economic impacts of the increase in tourism demand in the European food and accommodation industry from a cost perspective. This study contributes to the literature by addressing two key research questions. First, it explores the effects of increased tourism activities in Europe by assessing the total number of tourists, both domestic and international, and the average duration of their visits to a country, using the distance function approach to gauge the economic costs of increased tourism demand (Kumbhakar et al., 2015). Second, we investigate both the direct and indirect effects of increased tourism demand. In particular, we evaluate costs and distinguish overtourism phenomena based on the shape of the cost curve, that is, the declining or minimum costs indicative of favorable business outcomes and the increasing costs that signal unsustainable tourism due to congestion and pollution. We employ three alternative models; pooled linear regression, flexible nonparametric, and semiparametric smooth coefficient (SPSC), using data from 35 European countries from 2008 to 2020. Taking into account the multifaceted nature of tourism, we consider various accommodations and services to ensure a comprehensive analysis. Utilizing semiparametric modeling allows detailed exploration of heterogeneity in European tourism sectors. Furthermore, the economic impacts of tourism demand on the food and accommodation industries can vary between different countries in Europe. For example, popular tourist destinations can experience a higher concentration of tourism compared to lesser-known locations. This study goes beyond simply identifying trends; it provides insight into the economic implications of increased tourism demand and offers guidance for informed policy and business decisions at the national and regional levels.

This study provides a comprehensive understanding of the economic consequences of overtourism and solutions for mitigating these negative effects. This impact on economic sustainability is significant for businesses and regulators. As tourism demand surpasses sustainable levels, escalating costs and taxes can impede economic performance due to intensified competition and regulatory pressures. It has been argued that aligning operations with broader sustainability objectives and consumers’ behavioral nudges towards environmentally responsible choices can mitigate adverse economic impacts (Pulido-Fernández et al., 2019). This study’s novelty lies in its focus on the economic impacts from the perspective of business enterprises, providing insights for stakeholders and policymakers. It advocates for a balanced approach to tourism development, cognizant of its economic, environmental, and social dimensions, to foster sustainable growth in the sector.

Methodology and data

Methodology

Production economics offers various approaches for measuring productivity, including primal measures with a single output (a production function), primal measures with multiple outputs, and dual cost/profit/revenue functions (Kumbhakar et al., 2015). For measurement of tourism and hospitality productivity, either a single-output primal method (Assaf and Agbola, 2014; Chatzimichael and Liasidou, 2019) or the multiple-output Input Distance Function (IDF) approach (Abrate and Erbetta, 2010; Alemayehu and Kumbhakar, 2021) is used. The current study also applies an IDF approach with multiple outputs because it is difficult to obtain price data, service characteristics (Duncan, 1990; Gaynor and Anderson, 1995), and accurate cost measurement for the cost function approach. Previous studies have considered tourist arrivals (Assaf and Tsionas, 2018; Lozano and Gutiérrez, 2011) and duration of stay (Abad and Kongmanwatana, 2015; Assaf and Dwyer, 2013; Chatzimichael and Liasidou, 2019) as output.

However, this approach highlights that probabilistic demand for services influences cost functions, since demand is determined externally. Firms that accommodate fluctuating demand generate the probability of providing service rather than taking it as given (Duncan, 1990). That is, they must adjust their capacity and production levels to meet varying customer demands; and failing to do so results in under-utilization of capacity and challenges in minimizing costs. The costs associated with idle capacity and uncertain demand impact a firm’s cost structure through outputs, prices, and input utilization. Operational flexibility and adaptability are critical, which requires strategic planning, efficient resource management, and the ability to forecast and respond to demand fluctuations. This approach underscores the importance of maintaining reliable service despite uncertain demand and optimizing cost structures. Furthermore, tourism theory underscores tourism demand which is influenced by various factors, including income levels, relative prices, exchange rates, and external economic conditions (Witt and Witt, 1995), as well as seasonal variations, national tourism policies (Song and Witt, 2012), economic trends, international marketing efforts, and geopolitical stability (Li et al., 2005). Tourism demand significantly shapes the production process and cost structure within the accommodation industry. The impact of tourism demand on costs can be dissected into several factors, including the adjustment of room rates based on demand fluctuations, the increased use of variable inputs such as utilities and housekeeping during peak periods, and the optimization of capacity utilization and economies of scale as occupancy rates increase (Smeral, 2009). Efficient supply chain management is also crucial to meeting varying demand levels, which influences operational costs and infrastructure investments (Li et al., 2005). Thus, our methodology treats tourism demand as an environmental variable rather than a direct determinant of cost functions, offering the advantage of distinguishing between input-output variables and environmental factors such as tourism demand.

Thus, we begin with the transformation function F(x_it, y_it) = A_it, where x_it and y_it are vectors of input and output, respectively, and A is the neutral productivity shift function with unobserved productivity shock (random noise) and other exogenous shifters (if any). By imposing homogeneity of degree 1 (in x) restrictions on F(.), we can write it as an input distance function (IDF), that is, $\ln x_{1it} = f (\ln {\tilde{x}}_{jit}, \ln y_{mit}, t) + \ln A_{i t}$ , where ln x_1it is the natural log of one of the inputs (e.g., labor), ${\tilde{x}}_{i t}$ is x_jit/x_1it, j = 2, …, J and t is the time trend variable introduced to capture technical change. See (Kumbhakar, 2013) for details on the derivation of IDF from the transformation function. In our application, J = 3 and M = 7. We account for unobserved heterogeneity and dynamic adjustment using the lagged labor variable as an additional regressor, addressing potential biases in input coefficient estimation (Ackerberg et al., 2015; Gandhi et al., 2020; Levinsohn and Petrin, 2003). This technique ensures our analysis captures the dynamic nature of productivity and input interdependencies, leading to robust results.

Model 1: First, we consider the following log-linear IDF model in which ln A_it = β_tt + u_it:

\ln x_{1it} = β_{0} + β_{1} \ln x_{1it−1} + \sum_{j = 2}^{3} β_{j} \ln {\tilde{x}}_{jit} + \sum_{m = 1}^{7} γ_{m} \ln y_{mit} + β_{t} t + \sum_{k = 1}^{2} δ_{k} z_{kit} + u_{i t}

(1)

where β_j and β_t are the jth input elasticity and technical change, respectively; γ_m is the elasticity of the m^th output and u_it is the error term. Because the input distance function is dual to the cost function, these elasticity estimates can be interpreted as the cost elasticity. For example, the parameter δ_k captures the marginal effect of z_kit on the cost k = 1, 2. Although simple and intuitive, this model is restrictive because of the marginal effects of t and z_kit on the ln x_1it are constants for all tourism destinations, that is,

\frac{\partial \ln x_{1it}}{\partial t} = β_{t}

and

\frac{\partial \ln x_{1it}}{\partial z_{kit}} = δ_{k}

. Consequently, we extend the log-linear model in equations (1) to a more flexible nonparametric or semiparametric smooth coefficient (SPSC) form to allow for the heterogeneous marginal effects of z_kit.

Flexible marginal effects

To allow flexibility in marginal effects, we considered two models. In Model 2, we introduced a nonparametric function in which the z variables affect costs directly and indirectly unlike all other variables.

\ln x_{1it} = β_{0} (z) + β_{1} \ln x_{1it−1} + \sum_{j = 2}^{3} β_{j} \ln {\tilde{x}}_{jit} + \sum_{m = 1}^{7} γ_{m} \ln y_{mit} + β_{t} t + u_{i t}

(2)

where β₀(z) is a nonparametric function of z variables. Given that β₀(z) is nonparametric, the marginal effect of z_k is also nonparametric (fully flexible) and observation-specific.

We estimate the model in (2) in two steps. First, we obtain the conditional expectation of each of the covariates and the dependent variable and then use them to transform (2) to remove β₀(z). This is known as the Robinson transformation. Once the parameters (β_j are estimated, we obtain an estimate of β₀(z) in the second step. See Robinson (1988) for further details on the estimation steps.

Direct and indirect marginal effects

Our final model (Model 3) allowed flexible direct and indirect effects. The direct effects are the same as those in Model 2. Given that the z variables can also affect x₁ indirectly via covariates, Model 3 is specified as

\ln x_{1it} = β_{0} (z) + β_{1} \ln x_{1it−1} + \sum_{j = 2}^{3} β_{j} (z) \ln {\tilde{x}}_{jit} + \sum_{m = 1}^{7} γ_{m} (z) \ln y_{mit} + β_{t} (z) t + u_{i t}

(3)

where β₀(z), β_j(z), γ_m(z), and β_t(z) are unknown smooth nonparametric functions of z. This model is known as the SPSC model.

We employ Gaussian kernel functions and least-squares cross-validation, among the various kernel methods [see Henderson and Parmeter (2015) for more details] to estimate the nonparametric functional coefficients and compute the marginal effects. The Gaussian kernel balances bias and variance, ensuring robust estimation. Gaussian kernels capture data smoothness, offering compact support and noise reduction, enhancing accuracy. The estimation procedure for the SPSC model follows Li et al. (2002); Li and Racine (2010). First, we assume the intercept coefficient (β₀) and the slope coefficients (β_j and γ_m) as smooth nonparametric functions of Z. Second, we utilize this method to estimate these smooth functions, assigning weights based on the distance between observations and the point of interest. That means, the SPSC model provides flexibility for analyzing complex data relationships with nonparametric methods.

The bandwidth parameter, crucial for determining the degree of smoothing, is meticulously selected to strike a balance between bias and variance. The least squares cross-validation is chosen for its adaptability and ability to minimize prediction error, maintain model flexibility, and capture underlying relationships without undue bias. The bandwidth parameter is key, determined by least-squares cross-validation to balance bias and variance, as least-squares cross-validation minimizes prediction error, improving model performance. Third, we employ constrained estimation techniques to uphold constraints such as the non-negativity of marginal effects, thereby improving the economic interpretation of the estimated coefficients. Finally, we compute the marginal effects of Z on the outcome variable using the estimated smooth functions, allowing a complete analysis of their relationship. In essence, the SPSC model provides a flexible framework for scrutinizing intricate relationships in empirical data, providing nonparametric flexibility for modeling effects. If the functional coefficients are constant, Model 3 is reduced to Model 2. Similarly, Model 1 is a special case of Models 2 and 3.

Data

The data used in this study were obtained from the Eurostat database, which covers the period from 2005 to 2020 and comprises 576 observations from 37 European countries. However, after data cleaning, our final dataset consisted of 248 observations from 24 European countries, spanning 2008 to 2020. Several missing observations were excluded from the analysis using the software. This dataset offers detailed information on the production of the food and accommodation industry and the demand for tourism in different European Nomenclature (NACE) classifications, providing a broad temporal and geographical dimension.

The data comprise food and accommodation sectors at different levels. In this study, we considered level 2 classifications: I551, I552, I553, and I559, indicating hotels and similar, holidays and other short stay, camping recreation, and other accommodations, respectively. The food subsector includes I561, I562, and I563 for restaurants and mobile food, event catering and other food, as well as beverages and serving. As our empirical model analyzes multiple outputs, this study considers the outputs classified into these categories measured in monetary terms. The inputs are quantified in terms of labor, capital stock, and materials, following the standard model specification in the literature (Assaf and Barros, 2013). Labor is measured by wages and salaries, capital in gross investment, and materials in the purchase of goods and services not for resale. All these variables are also measured in millions of euros, except for gross investment, which was presented in 1000 euros and converted accordingly for the analysis. The time trend variable t is computed as the difference between the year and 2007, with the year ranging from 2008 to 2020.

The extant literature (e.g., Grönroos and Ojasalo, 2004; Smeral, 2009; Syverson, 2011) elucidates that labor inputs are assessed through labor services. Syverson (2011) articulates three methodologies for quantifying labor input in productivity analysis, each possessing distinct advantages and limitations. Smeral (2009) emphasizes the imperative to measure labor input in productivity analysis by quantifying labor quantity (e.g., hours worked) and considering factors that influence workforce quality and productivity. By integrating both quantitative and qualitative measures, economists can more accurately evaluate labor’s role in production processes and its contribution to overall productivity levels. The quantitative measure, such as the number of hours worked or the equivalent full-time employee, neglects the qualitative aspect of labor services. In contrast, the qualitative measure accounts for variations in skills, education, experience, and other factors that can influence the contributions of the workforce to the production process. Lastly, the wage bill disbursed to employees serves as a proxy for labor input, presuming that wage variations reflect differences in labor productivity and quality. Higher wages per employee are often perceived as employing more productive workers. Grönroos and Ojasalo (2004) advocate for this metric in the measurement of productivity in the service sector, although it may not comprehensively capture all dimensions of quality. Due to constraints on data accessibility, our study utilizes wage bills as a measure of labor input.

Tourism demand is measured by the number of tourist arrivals and average length of stay, with a distinction between domestic and international sources. These variables are measured in terms of the number of tourists and nights. The review by Syverson (2011) elucidates how economists assess and analyze productivity disparities using various data sources and methodologies. These data sources at various levels can capture distinctions between firms, industries, and countries. The study further underscores the potential to integrate these distinctions into productivity models through economic modeling. Since our research endeavors to comprehend the heterogeneity of productivity and the ramifications of tourism demand in Europe, it is situated within the realm of international comparisons that account for such heterogeneity using a semiparametric smooth coefficient model, which provides estimates for each observation. We believe that this captures the heterogeneity among the tourism industry in Europe more effectively, given the objective of the study.

Table 1 presents descriptive statistics for the variables used in the empirical analyses, offering insight into the importance of tourism to the economies of European countries. In particular, the sector pays wages and salaries of around four billion annually, invests about 13 billion, and spends about three billion on materials on average in the sample countries. The sector generates revenue of approximately 18 billion euros annually, with restaurants and mobile food services contributing the largest share (46%), followed by hotels and similar establishments (26%) and beverages and serving (13%). The dataset also reveals that 31 million tourists participate in tourism activities in these countries annually on average, with domestic tourists accounting for the largest share slightly above 60%. These tourists stay on average for three nights in these countries, foreigners stay longer than locals, and the sample countries are represented in seven out of thirteen years on average. Figure 1 illustrates the distribution of the sample countries over time.

Figure 1.

Sample distribution among European countries and years (2008-2020).

The temporal variability of the sample countries is important to capture the dynamic nature of economic and tourism developments in Europe. Bartelsman and Doms (2000) demonstrate that scrutinizing temporal variations within a country elucidates patterns, trends, and determinants of productivity fluctuations. Deciphering the causes of dispersion reveals the drivers of productivity with more precision. Analyzing technological adoption, regulatory changes, and human capital investments evaluates their long-term impacts. In addition, it elucidates industry dynamics, including firm entry and exit, ownership transitions, and mergers and acquisitions, which are crucial to comprehending competitive dynamics and productivity.

In summary, temporal variability is essential to our study, enhancing the robustness and credibility of our analyses by capturing diverse economic conditions and enabling sophisticated modeling techniques. This encompasses a spectrum of economic conditions, policy changes, and other determinants that impact productivity and tourism demand. By longitudinally observing a heterogeneous set of countries, our dataset significantly enhances the robustness of our analyses. Such variability mitigates biases inherent in static data and facilitates the application of advanced econometric techniques, such as the semiparametric smooth coefficient model, which yields precise estimates by accommodating heterogeneity. Consequently, our findings remain contemporaneous and reliable, thereby fortifying the validity of our conclusions.

Results

We analyze the impact of increased tourism on costs in the European Food and Accommodation industry using pooled linear regression, nonparametric, and SPSC models. The effects of tourist arrivals and length of stay (denoted z₁ and z₂) on costs are examined using three models. Model 1 represents the standard linear regression model, with z variables having constant marginal effects. In Model 2, the marginal effects are direct and non-parametric functions of z₁ and z₂. Model 3 extends this by incorporating both direct and indirect marginal effects, modeling all coefficients as non-parametric functions of z₁ and z₂.

The novelty of Models 2 and 3 lies in their ability to estimate the marginal effects of the z variables in a fully flexible (nonparametric) manner. In Model 2, the variables z directly influence the cost through the constant term (β₀), which is a nonparametric function of z₁ and z₂. In Model 3, the z variables affect the cost through all parameters, which are nonparametric functions of z. Thus, the marginal effect of z_k is obtained from $\partial \ln x_{1it} / \partial z_{kit} = \frac{\partial {\hat{β}}_{0}}{\partial z_{kit}}$ in Model 2. This can be interpreted as the percentage change in cost (multiplied by 100). In Model 3, the marginal effects were direct and indirect. That is, the z variables affect the cost directly via the constant term (β₀), as well as indirectly through all parameters β_j, γ_m, and β_t, as these parameters are nonparametric functions of z. These indirect effects are observation-specific and depend on specific variables z. Specifically, indirect effects include the impact of z_kit on ln x_1it through (i) input elasticity estimates, (ii) output elasticity estimates, and (iii) technical change parameter. We calculate the overall indirect marginal effect of z_k by adding these components.

{Indirect}_{k} = \frac{\partial {\hat{β}}_{1}}{\partial z_{kit}} \ln x_{i t - 1} + \sum_{j = 2}^{3} \frac{\partial {\hat{β}}_{j}}{\partial z_{kit}} \ln {\tilde{x}}_{jit} + \sum_{m = 1}^{7} \frac{\partial {\hat{γ}}_{m}}{\partial z_{kit}} \ln y_{mit} + \frac{\partial {\hat{β}}_{t}}{\partial z_{kit}}, t, k = 1,2 .

(4)

The overall marginal effect of z_k is the sum of direct and indirect effects.

\frac{\ln x_{1it}}{\partial z_{kit}} = \frac{\partial {\hat{β}}_{0}}{\partial z_{kit}} + \frac{\partial {\hat{β}}_{1}}{\partial z_{kit}} \ln x_{i t - 1} + \sum_{j = 2}^{3} \frac{\partial {\hat{β}}_{j}}{\partial z_{kit}} \ln \tilde{x} j i t + \sum_{m = 1}^{7} \frac{\partial {\hat{γ}}_{m}}{\partial z_{kit}} \ln y_{mit} + \frac{\partial {\hat{β}}_{t}}{\partial z_{kit}}, t, k = 1,2 .

(5)

We illustrate the statistical significance of the parameter estimates following Kumbhakar and Sun (2012) and Henderson et al. (2012). The criteria are as follows: the coefficient estimates are mapped along the 45° line, denoted by black stars, and the lower and upper 95% bootstrapped confidence intervals are illustrated in red and green, respectively. These bootstrapped confidence intervals were constructed using the standard formula, that is, coefficient estimates ± 2*standard error. The horizontal and vertical lines intersect at the center (mean) and divide the graph into four quadrants. An estimated observation is considered significant at the 5% level if it lies on the lower left or upper right side of the center; observations lying on the horizontal line are not significant.

The next subsection presents key findings on the direct, indirect, and overall impacts of the increase in tourism demand, derived from the SPSC model. The subsequent sections delve into point estimates of elasticity and returns to scale, while the discussion and implications section summarizes these findings and their practical implications.

Direct and indirect cost effects of increased tourism (z_k)

Table 2 and Figure 2 summarize the cost effects in the food and accommodation sector due to increased tourism. Direct effects average 0.29% [−0.57 to 0.95], while indirect effects are −0.27% [−0.97 to 0.59]. Overall effects average 0.10% [−0.00 to 0.03]. Positive direct costs and negative indirect impacts result in a minor overall negative impact. Thus, increased tourist arrivals raise direct costs but dampen indirect costs, with dampening effects prevailing.

Table 2.

Direct, indirect, and overall cost effects of increased tourism demand.

Variables	Tourist arrivals				Length of stay
Variables	Mean	25%	Median	75%	Mean	25%	Median	75%
DE	0.290	−0.566	0.266	0.950	0.460	−2.805	−0.038	3.24
DE	(0.080)	(0.030)	(0.049)	(0.080)	(0.370)	(0.161)	(0.255)	(0.43)
IE	−0.270	−0.973	−0.259	0.590	−0.79	−4.807	0.056	3.460
IE	(0.220)	(0.095)	(0.154)	(0.270)	(0.610)	(0.268)	(0.470)	(0.780)
OE	0.010	−0.004	0.005	0.030	−0.320	−2.308	−0.414	2.200
OE	(1.110)	(0.381)	(0.743)	(1.160)	(4.080)	(1.529)	(3.098)	(5.09)
Domestic vs. Foreign tourism arrivals
DE	4.934	−5.644	2.507	16.724
DE	(0.588)	(0.337)	(0.456)	(0.691)
IE	−4.986	−16.707	−3.194	5.872
IE	(1.419)	(0.700)	(1.075)	(1.766)
OE	−0.053	−0.459	−0.062	0.290
OE	(11.820)	(4.215)	(9.539)	(15.858)

Figure 2.

Density plots for cost effects of increased tourism.

Figure 3 and the classifications in Table 3 underscore the statistical significance of these effects, revealing a spectrum of positive, zero, or negative impacts, depending on the context. Specifically, the direct effects of tourist arrivals are positive in approximately 58% of observations and negative in approximately 37%. In contrast, indirect effects exhibit positive effects in approximately 44% of the cases and negative effects in approximately 46%. Overall, the effects are positive in about 64% of the cases and negative for the remaining observations.

Figure 3.

Statistical significance of cost effects of increased tourism.

Table 3.

Classification of direct, indirect, and overall effects.

	Positive	Negative	Not sign.
Tourist arrivals
Direct effects	0.582	0.371	0,047
Indirect effects	0.435	0.459	0.106
Overall effects	0.635	0.000	0.375
Length of stay
Direct effects	0.500	0.459	0.041
Indirect effects	0.506	0.459	0.035
Overall effects	0.453	0.100	0.447
Domestic vs. Foreign tourists
Direct effects	0.500	0.459	0.041
Indirect effects	0.371	0.565	0.100
Overall effects	0.441	0.012	0.982

The effects of length of stay are diverse. Direct effects average 0.46% [−2.81 to 3.24], and indirect effects average −0.78% [−4.81 to 3.46]. Overall, the average effect is −0.32 [−2.31 to 2.20]. Length of stay increases direct costs but reduces indirect costs, leading to lower overall costs. Longer stays result in higher direct costs but lower indirect costs. Table reftab:Table3 shows that 50% of observations report positive direct effects and 37% negative, while 51% show positive indirect effects and 46% negative. Overall, negative effects slightly prevail, representing 45% of the cases.

The proportion of domestic tourists affects costs. Direct cost effects were 4.93 [−5.64 to 16.72], while indirect effects were −4.99% [−16.71 to 5.87], resulting in an overall impact of −0.05 [−0.46 to 0.29]. Higher shares of domestic tourists increase direct costs, decrease indirect costs, and overall costs. Table 3 shows that 0.50% of direct effects are positive and 46% are negative; indirect effects are 0.37% positive and 56% negative. Overall, 44% of outcomes are positive, and 1.2% are negative.

Statistical significance of coefficient elasticity estimates and returns to scale estimates

Table 4 presents summary of the point estimates for the cost elasticity of input ratios and outputs and the returns to scale for the pooled linear regression, nonparametric, and SPSC models. The pooled linear model shows that all the coefficient estimates, including those of tourism demand, Z₁ and Z₂ are statistically significant at the level less than 5%. However, in the the nonparametric model, the output elasticity estimates are insignificant, but the lagged cost, as well as materials (x₃) are significant.

Table 4.

Elasticity estimates.

Variable	(1)	(2)				(2)
Variable	Mean	Mean	25%	50%	75%	Mean	25%	50%	75%
Intercept	0.059	5.954	5.019	5.796	6.498	2.237	1.607	2.425	2.827
Intercept	(0.309)	(0.281)	(0.248)	(0.270)	(0.304)	(0.048)	(0.003)	(0.004)	(0.057)
x ₁ _lag	0.019*	0.002				−0.006	−0.007	−0.006	−0.002
x ₁ _lag	(0.009)	(0.002)				(0.004)	(0.001)	(0.002)	(0.006)
x ₃	−0.655***	−0.054***				−0.817	−0.985	−0.806	−0.695
x ₃	(0.034)	(0.017)				(0.011)	(0.006)	(0.008)	(0.015)
y ₃	0.237***	0.000				0.076	0.046	0.071	0.104
y ₃	(0.032)	(0.012)				(0.007)	(0.004)	(0.006)	(0.010)
y ₄	0.061***	0.001				0.045	0.032	0.044	0.058
y ₄	(0.008)	(0.012)				(0.003)	(0.002)	(0.003)	(0.005)
y ₅	0.0257***	0.116				0.467	0.338	0.463	0.559
y ₅	(0.006)	(0.086)				(0.009)	(0.004)	(0.006)	(0.011)
y ₆	0.300***	0.046				0.189	0.097	0.147	0.270
y ₆	(0.012)	(0.014)				(0.016)	(0.009)	(10.020)	(0.020)
y ₇	0.107***	0.004				0.189	0.097	0.147	0.270
y ₇	(0.013)	(0.012)				(0.005)	(0.002)	(0.003)	(0.006)
t	0.020***	0.012				0.008	0.004	0.008	0.012
t	(0.003)	(0.004)				(0.001)	(0.001)	(0.002)	(0.001)
RTS	0.73%	†				79%
Z ₁	0.174***	1.183	−0.043**	0.041	0.280
Z ₁	(0.029)	(0.119)	(0.052)	(0.105)	(0.158)
Z ₂	0.152***	0.868	−1.562**	−0.228	0.230
Z ₂	(0.042)	(0.351)	0.152	0.263	0.516

Models (1), (2), and (3) present the results of linear regression, nonparametric, and Semiparametric Smooth Coefficient. Standard errors are in parentheses. ***, **, and * denote significance at the 0.1%, 1%, and 5% levels. † The output elasticity estimates are not statistically significant.

However, the effects of increased tourism demand (i.e., tourist arrivals and length of stay show contrasting effects with the results from the pooled OLS. The advantage of the SPSC model is that it provides the possibility of excluding outliers. Although some observations violate theoretical restrictions, this flexible modeling approach provides valuable insights into areas that require closer investigation. Figure 4 shows the density plot of these estimates, while Figure 5 presents the statistical significance and shows mixed results on the significance of these coefficient elasticity estimates; some point estimates are significant, while others are not. This variability is expected owing to the observation-specific nature of the SPSC model. These show heterogeneous results, with some negative, zero, or positive. RTS indicates scale economies, where increasing these outputs can lead to decreased costs (RTS < 1), constant cost changes (RTS = 1), or increased costs (RTS > 1). The results show that the RTS is 73% in the Pooled OLS and 79% [87, 99] in the Semiparametric model, on average. The estimates of technical change are similarly significant at less than the 0.1% level.

Figure 4.

Density plots of elasticity estimates.

Figure 5.

Elasticity estimates.

Discussion and implications

The European tourism industry is under increasing pressure due to the increasing demand for tourism. To gauge the impact of this surge on industry costs, we investigated the cost dynamics of the European Accommodation and Food sector using three models: pooled linear regression, nonparametric and semiparametric smooth coefficient (SPSC) models. Our findings show a nuanced picture of the factors influencing costs in this industry.

The direct effects of tourist arrivals and length of stay have heterogeneous effects on costs. Although direct effects leaned towards positive, indirect effects tended to be negative, resulting in a minor overall negative effect on average. Furthermore, the effects of tourist arrivals and length of stay varied in context, indicating nuanced impacts depending on specific circumstances. Furthermore, it was found that the proportion of domestic tourists influences costs, with higher proportions leading to lower direct costs but higher indirect costs on average.

These findings highlight the heterogeneity of the European tourism sector. In some countries, an increase in tourism demand can decrease or increase costs, depending on specific contextual factors. This heterogeneity in the impact of costs can be attributed to various channels or variables, such as tourist arrivals or length of stay. Higher tourism demand may increase direct costs but potentially decrease indirect costs on average, resulting in counterbalancing overall effects. This is in contrast to previous studies such as Larsen and Wolff (2019) and Incera and Fernández (2015) and Muštra et al. (2023) that found results in favor of domestic tourists relative to the international to increase tourist satisfaction, reduce income inequality, and improve economic resilience through cultural heritages.

Central to our investigation is an examination of the economic sustainability of European tourism from a cost perspective. By analyzing the cost impacts of heightened tourism demand, our objective is to provide empirical evidence on the industry’s sustainability. However, our findings provide a mixed picture. Tourism operations in some European countries are economically sustainable, while others are not, as illustrated in Figure 6 in the Appendix. This underscores the diverse nature of European countries in terms of their potential for tourism attraction and policy priorities, reflecting the complexities of the industry.

To explore the reasons for these heterogeneous effects, it is important to look into the differences in cost minimization efforts among industries and additional expenses associated with attracting and maintaining longer-staying tourists contribute to this complexity. To address these challenges, European tourism industries must adopt tailored strategies that align with their contexts and objectives. Industries facing declining costs should focus on increasing tourist stay, whereas those that operate optimally should maintain their operations. This study contributed to the evaluation of the cost impacts of increased tourism. Future research should extend this to a comprehensive analysis that includes profitability and reputation, which is crucial to understanding the broader implications of European tourism.

Overall, our study sheds light on the intricate relationship between tourism demand and cost of the European Food and Accommodation Industry. By uncovering the heterogeneous impacts and exploring the influence of domestic tourism, we offer valuable insights to industry stakeholders and policymakers. Targeted strategies should be implemented for mitigating cost pressures and ensure the long-term sustainability of European tourism.

Conclusion

This study explores the economic consequences of rising tourism demand on Europe’s food and accommodation sector, examining the impacts through various mechanisms and sources of tourism demand. We developed a country-specific tourism cost model utilizing the IDF methodology due to its cost interpretation – being a dual to the cost function, its ability to manage multiple inputs and outputs, and its independence from price data. We applied the model through pooled regression, nonparametric, and semiparametric smooth coefficients (SPSC), with data sourced from the Eurostat database over a span of approximately 13 years. The SPSC model is particularly flexible and beneficial as it addresses the variation in cost technologies and tourism potential among European countries, offering measures of both direct and indirect effects.

Our analysis showed heterogeneous impacts of tourist arrivals and length of stay on costs, with direct effects generally positive but indirect effects negative, resulting in a minor overall negative effect on average. Moreover, the study highlighted the nuanced impacts of tourist arrivals and length of stay depending on specific contexts, with the proportion of domestic tourists having an impact on costs.

These findings underscore the complexities within the European tourism industry, where increasing tourism demand can either decrease or increase costs depending on various contextual factors. This study also sheds light on the economic sustainability of European tourism operations, indicating a mixed picture across different countries. Although some demonstrate economically sustainable practices, others demonstrate unsustainable ones, reflecting the diverse nature of tourist attractions and policy priorities.

Furthermore, our investigation identified various mechanisms that contribute to these heterogeneous impacts, emphasizing the need for customized strategies aligned with specific contexts and objectives. By addressing these challenges, the European tourism industry can better navigate cost pressures and ensure long-term sustainability.

In summary, this study contributes to a broader understanding of the economic consequences of overtourism and offers insights into potential solutions for mitigating its negative effects. It is crucial for stakeholders and policymakers to implement targeted strategies based on these findings to address the challenges posed by the increasing demand for tourism while promoting economic sustainability.

Supplemental Material

Supplemental Material - Economic impacts of tourism demand on the European food and accommodation industry

Supplemental Material for Economic impacts of tourism demand on the European food and accommodation industry by Fikru K. Alemayehu, Subal C. Kumbhakar, and Gudbrand Lien in Tourism Economics.

Footnotes

Acknowledgements

We thank the editor and the two anonymous referees for their valuable comments on an earlier version of this paper. The usual disclaimer applies.

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 iDs

Fikru K Alemayehu

Subal C Kumbhakar

Gudbrand Lien

Supplemental Material

Supplemental material for this article is available online.

Author biographies

Fikru Kefyalew Alemayehu is an associate professor of Business Economics in the Department of Economics at the Inland Norway University of Applied Sciences. He received his Ph.D. in economics from the University of Stavanger, Norway, in 2020. His current research encompasses productivity and efficiency analysis, innovation and sustainability, and applied econometrics. He has published in high-impact journals, including Tourism Management, Technological Forecasting and Social Change, and Resource Policy, among others.

Subal C Kumbhakar is a distinguished research professor in the Department of Economics at the State University of New York. He is currently a coeditor of Empirical Economics and is serving on the Board of Editors/Associate Editors of nine international journals. He is a co-author of the Cambridge University Press textbooks A Practitioner’s Guide to Stochastic Frontier Analysis and Stochastic Frontier Analysis and the co-editor of the three-volume Springer Handbook of Produc- tion.

Gudbrand Lien is a professor in business economics at the Inland School of Business and Social Sciences, Inland Norway University of Applied Sciences in Lillehammer, Norway. He is also a senior researcher in agricultural economics at the Norwegian Institute of Bioeconomy Research (NIBIO) in Ås, Norway. Gudbrand is a co-author of the textbook “Coping with Risk in Agriculture,” and the edited book “Managing Risk in Agriculture” both published by CABI. He serves as a Board Member of Energies and the European Review of Agricultural Economics.

References

Abad

Kongmanwatana

(2015) Comparison of destination competitiveness ranking in the European Union using a non-parametric approach. Tourism Economics 21(2): 267–281.

Abrate

Erbetta

(2010) Efficiency and patterns of service mix in airport companies: an input distance function approach. Transportation Research Part E: Logistics and Transportation Review 46(5): 693–708.

Ackerberg

Caves

Frazer

(2015) Identification properties of recent production function estimators. Econometrica 83(6): 2411–2451.

Alemayehu

Kumbhakar

(2021) Excess capacity, production technology and technical inefficiency in hospitality sector. Tourism Management 82: 104202.

Allan

Connolly

Figus

, et al. (2022) Economic impacts of covid-19 on inbound and domestic tourism. Annals of Tourism Research Empirical Insights 3(2): 100075.

Assaf

Agbola

(2014) Efficiency analysis of the Australian accommodation industry: a Bayesian output distance function. Journal of Hospitality & Tourism Research 38(1): 116–132.

Assaf

Barros

(2013) A global benchmarking of the hotel industry. Tourism Economics 19(4): 811–821.

Assaf

Dwyer

(2013) Benchmarking international tourism destinations. Tourism Economics 19(6): 1233–1247.

Assaf

Tsionas

(2018) The estimation and decomposition of tourism productivity. Tourism Management 65: 131–142.

10.

Bartelsman

Doms

(2000) Understanding productivity: lessons from longitudinal microdata. Journal of Economic Literature 38(3): 569–594.

11.

Bramwell

Lane

(2000) Tourism Collaboration and Partnerships: Politics, Practice and Sustainability. Channel View Publications, Vol. 2.

12.

Brida

Lanzilotta

Pereyra

, et al. (2015) A nonlinear approach to the tourism-led growth hypothesis: the case of the mercosur. Current Issues in Tourism 18(7): 647–666.

13.

Capocchi

Vallone

Pierotti

, et al. (2019) Overtourism: a literature review to assess implications and future perspectives. Sustainability 11(12): 3303.

14.

Chatzimichael

Liasidou

(2019) A parametric decomposition of hotel-sector productivity growth. International Journal of Hospitality Management 76: 206–215.

15.

Corluka

(2019) Tourism seasonality–an overview. Journal of Business Paradigms 4(1): 21–43.

16.

Dritsakis

(2012) Tourism development and economic growth in seven mediterranean countries: a panel data approach. Tourism Economics 18(4): 801–816.

17.

Duncan

(1990) The effect of probabilistic demands on the structure of cost functions. Journal of Risk and Uncertainty 3(3): 211–220.

18.

Dwyer

Edwards

Mistilis

, et al. (2009) Destination and enterprise management for a tourism future. Tourism Management 30(1): 63–74.

19.

Figini

Patuelli

(2022) Estimating the economic impact of tourism in the European Union: review and computation. Journal of Travel Research 61(6): 1409–1423.

20.

Gandhi

Navarro

Rivers

(2020) On the identification of gross output production functions. Journal of Political Economy 128(8): 2973–3016.

21.

Garrigós-Simón

Galdón-Salvador

Gil-Pechuán

(2015) The economic sustainability of tourism growth through leakage calculation. Tourism Economics 21(4): 721–739.

22.

Gaynor

Anderson

(1995) Uncertain demand, the structure of hospital costs, and the cost of empty hospital beds. Journal of Health Economics 14(3): 291–317.

23.

Gössling

Peeters

Hall

, et al. (2012) Tourism and water use: supply, demand, and security. an international review. Tourism Management 33(1): 1–15.

24.

Gössling

Scott

Hall

(2020) Tourism and Water: Interactions, Impacts and Challenges. Channel View.

25.

Grönroos

Ojasalo

(2004) Service productivity: towards a conceptualization of the transformation of inputs into economic results in services. Journal of Business Research 57(4): 414–423.

26.

Henderson

Parmeter

(2015) Applied Nonparametric Econometrics. New York, NY: Cambridge University Press.

27.

Henderson

Kumbhakar

Parmeter

(2012) A simple method to visualize results in nonlinear regression models. Economics Letters 117(3): 578–581.

28.

Incera

Fernández

(2015) Tourism and income distribution: evidence from a developed regional economy. Tourism Management 48: 11–20.

29.

Kee

Chau

(2020) Economic sustainability of heritage conservation in Hong Kong: the impact of heritage buildings on adjacent property prices. Sustainable Development 28(1): 308–319.

30.

Kristjánsdóttir

Ólafsdóttir

Ragnarsdóttir

(2018) Reviewing integrated sustainability indicators for tourism. Journal of Sustainable Tourism 26(4): 583–599.

31.

Kumbhakar

(2013) Specification and estimation of multiple output technologies: a primal approach. European Journal of Operational Research 231(2): 465–473.

32.

Kumbhakar

Sun

(2012) Estimation of tfp growth: a semiparametric smooth coefficient approach. Empirical Economics 43(1): 1–24.

33.

Kumbhakar

Wang

Horncastle

, et al. (2015) A Practitioner’s Guide to Stochastic Frontier Analysis Using Stata. New York, NY: Cambridge University Press.

34.

Larsen

Wolff

(2019) In defence of the domestic tourist - a comparison of domestic and international tourists’ revisit-intentions, word-of-mouth propensity, satisfaction and expenditures. Scandinavian Journal of Hospitality and Tourism 19(4-5): 422–428.

35.

Levinsohn

Petrin

(2003) Estimating production functions using inputs to control for unobservables. The Review of Economic Studies 70(2): 317–341.

36.

Racine

(2010) Smooth varying-coefficient estimation and inference for qualitative and quantitative data. Econometric Theory 26(6): 1607–1637.

37.

Huang

, et al. (2002) Semiparametric smooth coefficient models. Journal of Business & Economic Statistics 20(3): 412–422.

38.

Song

Witt

(2005) Recent developments in econometric modeling and forecasting. Journal of Travel Research 44(1): 82–99.

39.

Lozano

Gutiérrez

(2011) Efficiency analysis of EU-25 member states as tourist destinations. International Journal of Services Technology and Management 15(1-2): 69–88.

40.

Muštra

Perić

BŠ

Pivčević

(2023) Cultural heritage sites, tourism and regional economic resilience. Papers in Regional Science 102(3): 465–483.

41.

Oklevik

Gössling

Hall

, et al. (2019) Overtourism, optimisation, and destination performance indicators: a case study of activities in Fjord Norway. Journal of Sustainable Tourism 27(12): 1804–1824.

42.

Peeters

Gössling

Klijs

, et al., 2021. Research for Tran committee-overtourism: impact and possible policy responses.

43.

Perkumienė

Pranskūnienė

(2019) Overtourism: between the right to travel and residents’ rights. Sustainability 11(7): 2138.

44.

Pulido-Fernández

Andrades-Caldito

Sánchez-Rivero

(2015) Is sustainable tourism an obstacle to the economic performance of the tourism industry? Evidence from an international empirical study. Journal of Sustainable Tourism 23(1): 47–64.

45.

Pulido-Fernández

Cárdenas-García

Espinosa-Pulido

(2019) Does environmental sustainability contribute to tourism growth? An analysis at the country level. Journal of Cleaner Production 213: 309–319.

46.

Qiu

Fan

Lyu

, et al. (2019) Analyzing the economic sustainability of tourism development: evidence from Hong Kong. Journal of Hospitality & Tourism Research 43(2): 226–248.

47.

Robinson

(1988) Root-n-consistent semiparametric regression. Econometrica: Journal of the Econometric Society 56: 931–954.

48.

Seraphin

Ivanov

(2020) Overtourism: a revenue management perspective. Journal of Revenue and Pricing Management 19(3): 146–150.

49.

Séraphin

Zaman

Olver

, et al. (2019) Destination branding and overtourism. Journal of Hospitality and Tourism Management 38(1): 1–4.

50.

Smeral

(2009) Growth accounting for hotel and restaurant industries. Journal of Travel Research 47(4): 413–424.

51.

Song

(2008) Tourism demand modelling and forecasting—a review of recent research. Tourism Management 29(2): 203–220.

52.

Song

Witt

(2012) Tourism Demand Modelling and Forecasting. New Delhi: Routledge.

53.

Song

Cao

(2018) Tourism and economic globalization: an emerging research agenda. Journal of Travel Research 57(8): 999–1011.

54.

Streimikiene

Svagzdiene

Jasinskas

, et al. (2021) Sustainable tourism development and competitiveness: the systematic literature review. Sustainable Development 29(1): 259–271.

55.

Syverson

(2011) What determines productivity? Journal of Economic Literature 49(2): 326–365.

56.

Vizek

Barbić

Časni

AČ

(2024) The impact of the tourism accommodation composition on housing prices: the case of Croatia. Tourism Economics 30(1): 267–274.

57.

Wall

(2020) From carrying capacity to overtourism: a perspective article. Tourism Review.

58.

Witt

(1995) Forecasting tourism demand: a review of empirical research. International Journal of Forecasting 11(3): 447–475.

59.

Zhu

Zhan

(2021) Is sustainable sevelopment reasonable for tourism destinations? An empirical study of the relationship between environmental competitiveness and tourism growth. Sustainable Development 29(1): 66–78.

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

0.08 MB

Economic impacts of tourism demand on the European food and accommodation industry

Abstract

Keywords

Introduction

Methodology and data

Methodology

Flexible marginal effects

Direct and indirect marginal effects

Data

Results

Direct and indirect cost effects of increased tourism (z k )

Statistical significance of coefficient elasticity estimates and returns to scale estimates

Discussion and implications

Conclusion

Supplemental Material

Supplemental Material - Economic impacts of tourism demand on the European food and accommodation industry

Footnotes

Acknowledgements

Declaration of conflicting interests

Funding

ORCID iDs

Supplemental Material

Author biographies

References

Supplementary Material

Direct and indirect cost effects of increased tourism (z_k)