What is the best month to visit Vietnam?
How Sunshine Duration Impacts Tourism in Vietnam: A 22-Year Climate-Tourism Study
Introduction
Vietnam’s tropical climate is a major draw for international travelers. In 2019 alone, the country welcomed over 18 million visitors. While warm weather generally encourages tourism, particularly from colder countries, excessive heat can deter potential travelers. Despite this dynamic, the specific role of sunshine duration in shaping tourism demand in tropical destinations like Vietnam has been largely overlooked. Most existing studies have focused on temperate regions, even as climate change begins to disrupt long-standing weather patterns that guide travel decisions.
This research investigates the relationship between sunshine duration and tourist arrivals in Vietnam over the period from 2002 to 2023. A time series regression model is used to capture both linear and nonlinear dynamics, incorporating seasonal, economic, and global shock variables. Prior studies such as Falk (2014) and Scott et al. (2004) suggest a delayed and sometimes non-linear relationship between sunshine and tourism. This paper builds on those insights with a focused empirical approach tailored to a tropical climate context.
Literature Review
Research in tourism consistently underscores the importance of weather, especially sunshine, as a driver of tourist activity. Gössling et al. (2012) emphasize that pleasant climatic conditions significantly enhance a destination’s appeal. Scott et al. (2004) found that while sunshine increases tourism in some regions, excessive heat can push visitors away. Falk (2014) used Austrian data to show that extended sunshine can increase domestic travel, but overly intense conditions may cause a reversal. Kim et al. (2019) observed similar trends in Jeju Island, where Chinese tourist numbers rose with sunshine, though regional factors like humidity had offsetting effects.
Vietnam, as a tropical country, presents a unique context where high levels of sunshine may shift from being attractive to oppressive. Yet local studies tend to prioritize economic factors over weather conditions. Truong and King (2009) focused on visitor satisfaction and expenditure, while Song and Li (2008) reviewed tourism demand without integrating climate variables. Kozak et al. (2008) employed time series methods in Turkey but did not control for economic shocks. This study addresses these gaps by modeling sunshine duration with both linear and quadratic terms and adjusting for inflation, exchange rates, seasonal events, and external disruptions.
Data
The dataset comprises monthly observations from January 2002 to December 2023, totaling 264 data points. The dependent variable is tourist arrivals, sourced from CEIC, while sunshine duration and economic indicators are obtained from the General Statistics Office of Vietnam (GSO). Sunshine duration is averaged across 15 weather stations in Vietnam. One missing observation in November 2003 was imputed using the average of the remaining stations.
To test for diminishing returns of sunshine, a quadratic term is created by squaring the sunshine duration. Control variables include the Consumer Price Index (CPI) and a USD Index to measure inflation and exchange rate effects. Seasonal dummies capture known peaks, including Tet (Lunar New Year), the dry season (November to April), and Christmas.
Although the dataset provides broad coverage, it does have limitations. It averages sunshine across all regions, ignoring local variations in tourism and climate. Additionally, political and economic events outside of Vietnam, such as regional conflicts or financial crises, are not explicitly included.
Method
Four models are tested. The first regresses log tourist arrivals on sunshine duration, seasonal dummies, and economic variables. The second introduces a squared sunshine term to test for non-linear effects. The third adds a lagged dependent variable to address serial correlation. The fourth removes log(CPI) to reduce multicollinearity with the USD Index.
Each model is tested for serial correlation using the Durbin-Watson test, for heteroscedasticity using the Breusch-Pagan test, and for stationarity using the Augmented Dickey-Fuller test. Variance Inflation Factor (VIF) scores are computed to assess multicollinearity.
Results and Discussion
The first model shows a small, statistically insignificant effect of sunshine duration. However, introducing a squared term in Model 2 yields significant results: the linear term is positive, and the squared term is negative, confirming that too much sunshine can reduce tourism. This supports Scott et al. (2004)’s findings on diminishing returns.
Model 3, which adds a lagged dependent variable, significantly improves model fit. Tourist arrivals in one month are strongly influenced by the previous month’s numbers, with a coefficient of 0.728. Seasonal dummies lose some significance, indicating that much of the variation is captured by the lag.
Model 4 removes log(CPI), improving the significance of the USD Index. The optimal level of sunshine is around 167.1 hours per month, beyond which additional sunshine negatively affects tourism. The model is robust to serial correlation and stationary, though heteroscedasticity persists.
Conclusion
The analysis finds that sunshine duration plays a statistically significant but nonlinear role in Vietnam’s tourism demand. Moderate sunshine increases tourist arrivals, but excessive sunshine leads to declining interest. The optimal level of sunshine is approximately 167 hours per month. Tourist behavior is also influenced by previous travel trends, with strong month-to-month persistence.
While economic variables show limited impact, likely due to multicollinearity, the USD Index becomes more meaningful when CPI is removed. The study’s limitations, including regional averaging and unmodeled external shocks, suggest opportunities for further research using panel data and location-specific variables.
This study contributes to understanding how tropical climates interact with global tourism flows and offers valuable insights for destination managers and policymakers seeking to adapt to changing climate patterns.
