Golf simulation, known as "screen golf" in Korea, has rapidly expanded due to technological advancements and changing consumption patterns. It offers accessibility and affordability compared to traditional golf courses, attracting a broader user base and contributing to the economy through employment and revenue. In 2016, amidst economic weakness, South Korea actively promoted the golf simulation industry. While the industry's economic impact is significant, empirical research in this area remains limited. This study addresses this gap by analyzing the impact of golf simulation on Korean economic growth (2000-2020), considering its effects on employment, added value, and other sports-related industries. The SVAR model is employed to understand the complex interrelationships between these factors.

Existing literature extensively examines various aspects of the golf industry, including sporting skills, golf course management, golf tourism, and environmental impacts. Studies by Gelan (2003), Poulin et al. (2006), and Zhang et al. (2016) discuss the economic benefits of golf leagues, the challenges faced by golf companies, and product specialization. Research on golf tourism explores the economic effects and potential environmental concerns (Brown et al., 2006; Petrick and Backman, 2002; Tassiopoulos and Haydam, 2008; Kim et al., 2005; Hodgkison et al., 2007; Tanner and Gange, 2005; Hwang, 2017). Min and Kim (2018) and Bae et al. (2010) focus on technological advancements in golf simulation. Ko (2015) studies the economic effects of golf in Korea, but lacks systematic quantitative analysis of the shift from outdoor to indoor golf. This study bridges this gap by providing a comprehensive, quantitative analysis using the SVAR model.

The study employs an SVAR model to analyze the dynamic relationships between four variables: GDP per capita (LNPGDP), labor force (LNL), sports industry investment (LNSP), and the golf simulation industry (LNGF). A unit root test using the Augmented Dickey-Fuller (ADF) test is conducted to confirm stationarity. Data from 2000 to 2020 from sources including the Korea Sports Industry Report, Korea National Statistical Office, and the World Bank are used. The SVAR model is estimated using two types of constraints: short-term and long-term. Short-term constraints are imposed based on economic theory, assuming that the current period's golf simulation industry is not directly affected by contemporaneous changes in other variables. Long-term constraints are based on the Blanchard and Quah (1988) approach, focusing on long-run relationships between variables. Impulse response functions (IRFs) and variance decomposition analysis are utilized to analyze the dynamic effects and contributions of each variable to the overall system. Cholesky decomposition is used for structural identification. The stability of the SVAR model is checked using the inverse roots of the AR characteristic polynomial. Johansen Cointegration test is used to verify existence of long term relationships. The optimal lag order for the SVAR model is determined using several information criteria (LR, FPE, AIC, SIC, and HQIC).

Unit root tests indicate that all variables are non-stationary at their original levels but stationary after first-order differencing. The optimal lag order for the SVAR model is 1, and the model is stable. The Johansen cointegration test reveals two cointegrating relationships, suggesting long-term equilibrium relations between the variables. Short-term impulse response analysis reveals a strong correlation between the golf simulation industry, sports industry investment, and the labor population. However, the relationship with economic growth is weak and uncertain. Long-term impulse response analysis shows a unidirectional positive impact from golf simulation, sports investment, and the labor force on economic growth, but the effect is less volatile. Variance decomposition shows that the golf simulation industry explains a significant portion of the variance in its own fluctuations, while also contributing positively to the variance of economic growth. The labor force and sports industry investments are found to have positive but less significant contributions to the variance of economic growth. In the long run, a positive shock to economic growth leads to lasting positive impacts on sports industry investment and economic growth itself. The impact on the labor force is less certain.

The findings suggest a strong short-term relationship between the golf simulation industry and the labor market and sports industry investments. The long-term positive relationship between the golf simulation industry, sports industry investments, labor force, and economic growth highlights the industry’s contribution to the Korean economy. The relatively weak short-term impact of the golf simulation industry on economic growth may be due to the time lag required for the industry to fully translate into macroeconomic effects. The “U-shaped” market concentration of the golf simulation industry, and the high concentration of labor and investment in developed regions, contribute to regional disparities. The study supports the view that the golf simulation industry fosters economic growth, particularly in the long term and in relation to employment and sports-related investments.

This study provides quantitative evidence of the positive economic impact of the golf simulation industry in Korea. Both short-term and long-term analyses indicate positive contributions to employment and economic growth, albeit with some regional disparities. The findings underscore the importance of government support and strategic development of the industry. Future research could explore the industry's environmental impact and further examine regional variations in its economic effects. Additional research could also focus on other simulated sports industries to establish broader implications.

The study uses annual data which limits the granularity of the analysis. It focuses solely on Korea and might not be generalizable to other contexts. The SVAR model relies on specific assumptions about causality and may not capture all the complex interactions within the economy. The model's reliance on aggregate data might obscure heterogeneity within the golf simulation industry or specific regional dynamics.