The relationship between out-of-pocket health expenditures and outbound health tourism

This study systematically examines the impact of out-of-pocket health expenditures on outbound medical tourism using panel data covering 80 countries over the period 2005– 2022. Outbound medical tourism is analyzed through two alternative operationalizations: per capita outbound medical tourism expenditure (Sturizm2, absolute level) and the share of outbound medical tourism expenditures in total service trade (Sturizm21, relative share). Out-of-pocket health expenditures are captured using three different indicators: their share in total health expenditures, per capita values in current U.S. dollars, and per capita values adjusted for purchasing power parity (PPP). Methodologically, the study employs fixed- effects panel data models to control for unobserved, country-specific heterogeneity. To ensure robustness against cross-sectional dependence, serial correlation, and heteroskedasticity, Driscoll–Kraay robust standard errors are used. Based on a total of 29 different model specifications, the empirical results reveal the presence of strong dynamic persistence (path dependence) in outbound medical tourism. The coefficients of the lagged dependent variable are positive and highly statistically significant across all models, indicating substantial inertia. Specifically, the estimated coefficients range between 0.49 and 0.56 for per capita outbound medical tourism expenditures (Sturizm2) and between 0.43 and 0.52 for the relative share measure (Sturizm21) (p < 0.01). Per capita income, representing the level of economic prosperity, exhibits a positive and robust effect on absolute outbound medical tourism expenditures (Sturizm2), while its effect on the relative share indicator (Sturizm21) is statistically insignificant. This finding suggests that higher income levels primarily translate into increased spending volumes rather than changes in sectoral specialization. The price level variable produces measurement-dependent effects: in per capita expenditure models, positive and significant coefficients point to a "premium segment" effect, reflecting demand for higher-cost yet higher-quality services, whereas in relative share models (Sturizm21), the price effect weakens or becomes statistically insignificant. The core contribution of this study lies in demonstrating that the effect of out-of-pocket health expenditures on outbound medical tourism is not unidirectional, but rather multidimensional and context-dependent, varying across time horizons and measurement approaches. From a long-run structural perspective, higher average out-of-pocket expenditure burdens are found to negatively affect outbound medical tourism performance, as evidenced by country-mean models. This result reflects weaknesses in financial protection mechanisms within health financing systems and indicates increased financial risk exposure for individuals seeking healthcare services. In contrast, short-run dynamic models reveal that annual increases in out-of-pocket health expenditures exert a positive effect on outbound medical tourism, likely reflecting private sector expansion and market dynamism. Moreover, in models using the relative share indicator (Sturizm21), out-of-pocket expenditures generally display positive coefficients, suggesting that health systems with a stronger reliance on private financing provide a more favorable environment for the commercialization of medical tourism. Indicators representing health system capacity—namely hospital bed density, physician density, and nurse density—are mostly found to be statistically insignificant. This finding implies that outbound medical tourism decisions are driven less by quantitative capacity measures and more by qualitative factors, such as perceived service quality, specialization, technological sophistication, and institutional trust. The dummy variable capturing the COVID-19 pandemic yields consistently negative and statistically significant coefficients across all model specifications, confirming a substantial decline in outbound medical tourism during the pandemic period. Overall, the study demonstrates that the choice of the dependent variable critically shapes policy-relevant conclusions in the context of medical tourism. While economic development and price dynamics are more influential in increasing per capita outbound medical tourism expenditures (Sturizm2), private sector dynamism and market structure play a more decisive role in raising the relative importance of medical tourism within service trade (Sturizm21). The findings highlight that a balanced combination of strong public financing and effective private sector participation is essential for achieving sustainable medical tourism performance. Furthermore, the results underscore that measurement choices systematically influence empirical outcomes, emphasizing the importance of methodological precision in future research on medical tourism and health economics. Keywords: Health tourism, Out-of-pocket health expenditures, Health financing, Panel data analysis, Fixed effects model.