Modeling the link between tourism and economic development: evidence from homogeneous panels of countries

Tourism's global significance is undeniable, exceeding the growth rates of other economic sectors and contributing substantially to global GDP and employment. While the relationship between tourism and economic growth has been extensively studied, the link between tourism and broader economic development remains less explored. Many countries view tourism as a tool for economic growth and socio-economic advancement, impacting factors like education, safety, and health, beyond mere GDP increases. International organizations actively fund tourism projects, aiming to improve socio-economic conditions, particularly in less developed nations. This study addresses the gap in the literature by empirically analyzing the causal relationship between tourism and multi-dimensional economic development, encompassing economic and socio-economic variables. The research employs panel data from 123 countries between 1995 and 2019, using the Dumitrescu and Hurlin adaptation of the Granger causality test. A crucial aspect of the study is its critical examination of using heterogeneous panels in causal analyses, comparing results from the entire panel with those obtained from homogeneous country groups based on tourism specialization and development levels.

Existing literature on the tourism-economic development nexus is limited, with studies often focusing on specific countries or regions. While numerous organizations recognize tourism's developmental potential, rooted in the UNWTO's Manila Declaration, empirical analyses remain scarce. Some studies have explored the causal relationship using heterogeneous samples without prior grouping, leading to potentially misleading generalizations. Other studies have addressed this limitation by analyzing homogeneous groups, yielding varying results: some find a bidirectional relationship, others a unidirectional relationship (tourism impacting development or vice versa), and some report no relationship. The inconsistency underscores the need for a more nuanced approach considering country-specific contexts.

This study employs panel data from 123 countries spanning 1995-2019. Tourism is measured using the Tourist Specialization Rate (TIR), calculated as international tourists per capita, obtained from UNWTO and World Bank data. Economic development is measured using the Human Development Index (HDI) from the UNDP. The study utilizes the Dumitrescu and Hurlin (2012) Granger causality test, which accommodates heterogeneity in panel data, unlike standard Granger causality tests. The analysis is conducted both on the full panel and on homogeneous country groups identified based on their tourism specialization and development levels (using the classification from Brida et al., 2023). Before testing for causality, the study addresses cross-sectional dependence using the Pesaran (2004) test and ensures variable stationarity using the Pesaran (2007) CIPS test. Bootstrap techniques with 500 replications are used to correct p-values for cross-sectional dependence. The analysis focuses on the first differences of the variables to ensure stationarity, thus examining the growth rates of tourism specialization and development.

The Pesaran cross-sectional dependence test confirms significant dependence between the variables across all countries and within the homogeneous groups. The CIPS test indicates that both the log-transformed variables and their first differences are stationary. The Dumitrescu and Hurlin causality test reveals a one-way causal relationship from tourism to development for the entire sample (with a three-lag model), suggesting that tourism positively impacts development. However, this general conclusion is not valid for all country groups. When analyzing homogeneous groups, this causal relationship (tourism to development) is only significant for Group C (countries with low tourism specialization and low development levels). Conversely, the causality test for the effect of development on tourism shows no relationship for the whole sample. However, for Group A (countries with high tourism specialization and high development), a significant one-way relationship exists where high development levels positively influence tourism specialization levels. This suggests that in countries with high development levels, existing socioeconomic conditions contribute to increased tourist arrivals.

The findings challenge the simplistic notion that tourism universally contributes to development. The results highlight that the impact of tourism on development is contingent on pre-existing conditions. In countries with low tourism specialization and low development, increased tourism can contribute to improved development, although further investment is needed to fully realize its potential. In contrast, high levels of existing economic development attract tourists, reinforcing the link between development and tourism specialization. The study's heterogeneous panel analysis reveals the pitfalls of generalizing conclusions drawn from large, diverse samples. A one-size-fits-all policy approach to tourism development is not suitable; instead, tailored strategies are needed considering country-specific characteristics.

This research fills a gap in the literature by providing a nuanced analysis of the tourism-development relationship. The findings demonstrate the conditional nature of this relationship, dependent on initial levels of tourism specialization and development. For low-specialization, low-development countries, investment in tourism infrastructure and market positioning is crucial to harness its development potential. In contrast, for high-development countries, continued investments in infrastructure, security, and other socio-economic factors will further enhance tourism competitiveness. Future research could explore regional-level analyses and incorporate additional country-specific factors to enhance the understanding of the intricate tourism-development nexus.

The study's national scope might overlook the regional and local variations in tourism impacts. Future research could explore subnational levels. The analysis focuses on a specific set of variables; other factors might also play significant roles in mediating the tourism-development relationship, which could be considered in future studies.