The Effect of Russia - Ukraine war on energy sector shares

The wars that have occurred and are currently occurring between nations have led to global impacts. As a result of these wars, not only are there human casualties, but serious global effects are also triggered. These effects not only reduce the standard of living but also cause significant losses in economic, financial, food supply chains, and energy sectors. The war between Russia and Ukraine, which started in 2022, has not only affected these two countries but has also harmed other nations through its negative consequences. Russia and Ukraine are key players in the global food supply and energy distribution chains. The tensions between these two nations have led to global crises. Russia is a country with a major share in global energy resources, while Ukraine plays a critical role in energy distribution to other countries. The war has caused disruptions in energy distribution, leading to supply issues. The purpose of this study is to reveal the effects of the Russia-Ukraine war on the energy sector stocks. In the study, based on G20 countries, the closing data for energy stocks of three companies selected from the largest energy stocks of each country during the period from January 2022 to July 2024 were collected. Additionally, industrial production rates of G20 countries for the same period were included. Analyses and findings based on these data were interpreted. According to the research, there are 57 stock indices, with 19 countries and three companies from each. These stocks are grouped as SH1, SH2, SH3, and SU. In the analysis section, panel data analysis was initially applied. For SH1, SH2, and SH3, all results rejected the null hypothesis of no cross-sectional dependence (H0), indicating the presence of cross-sectional dependence in the series. Similarly, except for the CDLM test, all other tests demonstrated that the SU variable also contains cross-sectional dependence. Based on these results, second-generation tests were utilized in subsequent analyses. The Hsiao homogeneity-heterogeneity test was applied to the dataset, and the findings were presented in table form in Section 3. According to these findings, based on the results of the H1, H2, and H3 hypotheses, all variables constituting the dataset were found to be heterogeneous. According to the CIPS test results, SH1, SH2, and SH3 variables contain unit roots at their level values, indicating that these series are I(1). On the other hand, since the null hypothesis for the SU variable could not be rejected, it was determined that SU contains a unit root at its level value, meaning that the SU variable is I(0). Since the SU variable is stationary at the level, a cointegration test was not applied to the dataset. At this stage, the most significant analysis that can be performed is causality analysis. Considering this, the Toda-Yamamoto causality analysis developed by Emirmahmutoğlu and Köse (2011) was preferred. Based on the results of the analysis, a unidirectional Granger causality from SH1 to SU was found to be statistically significant at a 5% significance level. However, the reverse was not valid. It was observed that there is bidirectional Granger causality between SH2 and SU, significant at a 5% level. Consequently, it was concluded that SH2 and SU have a bidirectional causal relationship. No Granger causality was observed between SH3 and SU in either direction. As a result, no Granger causality was found between SH3 and SU.