This study employed monthly time series data. The secondary data was ­collected from January 1999–October 2022 (286 data observations). Table 1 shows the variables used in this study were the index of EU19 total food imports from Russia (as dependent variable), inflation, real exchange rate, consumer confidence index, money supply, unemployment rate, the index of EU19 total food imports from the U.S., the index of EU19 total food imports from China, crude oil prices, dummy recession, and dummy EU sanction (as explanatory variables).

The empirical analysis begins with a unit root test or the stationarity test before the estimation. The stationarity test was performed to eliminate spurious regression caused by using nonstationary time-series data throughout the period. One type of test is used to evaluate the stationarity of the variables, including Augmented Dickey–Fuller (ADF) (Wooldridge, 2020).

The two-stage least squares (TSLS) was used to analyze all variables to address­ an endogeneity issue, especially the INF_t variable. Endogeneity occurs since INF_t is supposed to influence EU19 food imports while other variables also influence INF_t at the same time (Prasada et al., 2022). The TSLS model is solved in two steps: addresses the INF_t’s endogeneity problem and determines the factors­ in­fluencing EU19 food imports from Russia (Greene, 2003). The following equations are used in each step:

First step:

The function estimates the statistical relationship between INF_t and its determinant factors:

INF_t = β₀ + β₁RER_t + β₂MON_t + β₃CCI_t + β₄UNE_t + μ. (1)

Second step:

The function estimates the statistical relationship between EU19 total food, beverages, drinks, tobacco, live animals, and animal and vegetable oils, fats and waxes imports from Russia and its determinant factors:

RUS_t = γ₀ + γ₁INF_t + γ₂US_t + γ₃CHI_t + γ₄OIL_t + γ₅REC_t + γ₆SAN_t + σ. (2)

The TSLS model must pass several post-estimation tests to be valid. Post-estimation tests for the TSLS model include: (1) the model must have an endogeneity problem (Li et al., 2021); (2) the model’s instrument variables are strongly correlated with endogenous regressors (Choi et al., 2018), and 3) the TSLS model must meet the identification restriction test criteria (Mariano, 2007).

The results of a dummy variable analysis from the TSLS model can only be used to determine whether there are differences in EU19 food imports from Russia following the recession and the implementation of EU sanctions. As a result, the Propensity Score Matching (PSM) method must be used after the TSLS analysis (Kuss et al., 2016).

ATT = E (R₁ | I = 1) – E (R₀ | I = 0), (3)

ATT = E{R₁ | I = 1, p (Z)} – E{R₀ | I = 0, p (Z)}, (4)

where ATT (average treatment effect of the treated group) represents the impact of implementing the policy; I is the indicators of the recession and the implementation of the EU sanction policy (I = 0 control group, I = 1 treatment group: the recession and the EU sanction treatment); R₀ and R₁ show the outcome value of control data and from treatment data; p (Z) is the propensity score. p (Z) is obtained from the probit estimation of the recession and the EU sanction.

Before the PSM analysis results can be properly interpreted, several post-estimation stages must be completed. If the PSM meets two basic assumptions, it is valid: conditional independence and overlapping assumptions (Sseguya et al., 2021). Furthermore, the PSM model is prone to bias due to factors not observed in the model (Mavromaras et al., 2009). Therefore, the PSM output must be re-analyzed with a sensitivity test. The Mantel–Haenszel bounds sensitivity test method was used in this study, which has advantages for data analysis that focuses on binary-outcome variables (Becker and Caliendo, 2007).

First of all, we performed the ADF unit root test to determine the stationarity of the data. Unit root test shows that only RUS_t is stationary at level. At the same time, INF_t, RER_t, MON_t, CCI_t, UNE_t, US_t, CHI_t, and OIL_t are stationary at the first-difference level (Table 2).

The TSLS model was used to analyze all variables after the data became stationary. Several post-estimation tests were performed on the TSLS model to determine whether it is suitable for determining the factors influencing EU19 food imports from Russia (Table 3). The model shows that the endogeneity test produces a Hausman of 4.85200 at the significance level, indicating that the equation has an endogeneity problem. The overidentification test and the weak instrument test show a significant value at the 5% alpha level, meaning that the structural model is included in the overidentified category (the Sargan statistic is 15.83650) and each equation has a strong instrument variable (Stock and Yogo statistic value of 12.92740).

According to our studies, the following explanatory variables influence the INF in the EU19: consumer confidence index (CCI) and unemployment rate (UNE). Increases in both explanatory variables reduce the INF in EU19. Consumer confidence reflects how consumers assess their financial ability, purchasing habits, and overall economic condition (Shayaa et al., 2018). According to Ekren et al. (2017), CCI can reduce inflation in Europe. The same thing happened in China, where inflation gradually fell after the CCI surpassed 100 (Wang and Li, 2012). Optimism about market conditions leads to increased expenditures on goods production by producers and firms (Sinamo and Hanggraeni, 2022). Investors will be eager to fund various businesses and FDI inflows will rise under this condition (Verma and Bansal, 2021). These various activities encourage increased domestic production, lowered product prices, and reduced inflation.

An increase in unemployment leads to a decrease in inflation in the EU. The findings­ of this analysis are consistent with the Phillips curve model, which states that unemployment and inflation have a negative relationship (Liargovas and Psychalis, 2022). Our findings are even more extreme, with inflation falling by 1.354% as the unemployment rate rises by 1%. However, conditions in the EU can change quickly because the relationship between these two variables tends to be reversed (Liargovas and Psychalis, 2022). Phelps and Friedman stated that an unemployment rate above the natural limit will cause an increase in inflation (Popescu and Diaconu, 2022).

The RER of a country is a key indicator for assessing its trade capabilities and current import/export situation. EU countries keep the RER stable to maintain the current account balance, accelerate economic competitiveness, and encourage exports (Nikas et al., 2019). Hence, the euro has low volatility and the amounts are well regulated, so the RER and MON have no impact on INF. The EU’s monetary policy also promotes a stable relationship between money and prices as a precondition for the optimum monetary aggregates condition (Jung and Carcel Villanova, 2020).

According to our findings, an increase in INF, US, CHI, and REC will lead to an increase in EU19 food imports from Russia, whereas OIL and SAN have no impact on EU19 food imports from Russia. As consumer prices or inflation rise, domestic products become more expensive than imported products. Hence, imported products will more easily enter a country and be liked by consumers. According to a study by Černý et al. (2021), import demand is indeed highly and positively correlated with inflation in the EU. Inflation in the EU has been low and stable thus far, causing import volumes to fall. This situation changed when inflation began to rise, causing imports to increase dramatically (Ben Cheikh and Rault, 2017).

During the study period, EU19 food imports from Russia increased (Fig. 1). Russia is the world’s leading producer of several food commodities, the fourth-largest producer of wheat, the eighth-largest producer of soybeans, and the tenth-largest producer of maize (Nasir et al., 2022). Russian agriculture has shown stable growth since 1999 after a significant decline in the early 1990s and the long process of transformation. The food trade balance is steadily improving and the share of imported food in retail markets is decreasing due to the government’s import substitution policies (Uzun et al., 2019). Several conditions and domestic policies contributed to this result. Russia can increase cropland area and production efficiency; distribute direct subsidies for fertilizers, fuel, lubricants, and soil nutrients; subsidize interest rates on agricultural loans and insurances; develop supply chain infrastructure and market access; and improve the knowledge and skills of farmers (Tleubayev et al., 2022).

Despite its strong performance, Russian food exports to the EU19 remain lower than those of China and the U.S. The US and China can dominate the global food trade, including exporting to the EU. In addition, both countries produce more food than Russia. China is the second-largest wheat producer globally, the fourth-largest soybean producer, and the second-largest maize producer. Similarly, the U.S. is the world’s fifth-largest wheat producer, the world’s second-largest soybean producer, and the world’s largest maize producer. China and the U.S. have a comparative advantage in global food trade, both due to low product prices, reliance on intellectual property rights, and product brands (Nasir et al., 2022). However, Russia is closer in distance to the EU so transportation costs are lower than the U.S. and China. Hence, the U.S., China, and Russia have their advantages, and their food exports to Europe are all increasing.

Trade relations between the EU19 and China or the U.S. have flourished in recent decades. China and the U.S. are engaged in a „silent war“ for global economic hegemony. Both countries are also involved in direct conflict over the food trade. As a result, China and the U.S. are looking for new markets for their food products and the most potential target is the EU. This is reflected in the EU and US commitments to bilateral trade agreements such as the Transatlantic Trade and Investment Partnership (TTIP). This agreement gives American companies greater access to European markets and equalizes perceptions of quality and food safety standards between the U.S. and EU (Pietrzyck et al., 2021). The Chinese and the U.S. governments also provide subsidies to farmers to increase agricultural competitiveness. The competition between both countries has allowed Russia to enter the European market.

The economic recession, which we use as dummy variables in this study, has a significant effect on EU19 food imports from Russia. The economic recession did not harm EU food imports from Russia. Food businesses in Russia have successfully chosen strategic development paths such as focusing on the needs of the most promising client groups, expanding service offerings, and expanding geographically (Dybskaya and Vinogradov, 2018). Meanwhile, the per capita income of EU countries increased from 1999 to 2022, causing imports to rise despite the economic downturn.

Next, we performed a PSM analysis to examine how REC affected changes in the import value of RUS. The results of the balance test show a reduction in the mean absolute standard bias between before and after matching. The decrease in the total bias is 99.50%, indicating that the impact evaluation results were unbiased (Table 4).

The overlapping assumptions are met before and after the EU recession (Fig. 2). All data can be perfectly matched, allowing for the proper use of research data to assess the impact of a policy or condition. These findings indicate that the matching quality was well maintained.

The PSM analysis produced consistent results with the TSLS analysis. The PSM test is also significant (t-statistics = 2.56) and unbiased (MH Bounds sensitivity­ = 4.56), meaning that the remaining unobserved individual heterogeneity after applying the PSM model is not a problem. The PSM analysis results show that EU19 food imports from Russia continued to increase despite the EU currently being in recession. EU19 food imports from Russia increased by 36.74 units after the EU experienced recession (Table 5).

According to this study, oil prices do not affect EU19 food imports from Russia. Volatility in global oil prices will not continue to have an impact on global food prices. Olayungbo (2021) found there is a unidirectional causality relationship originating from food prices to oil prices in the long run; however, there is no reverse relationship between oil prices and food prices. This implies that changes in oil prices can be explained by the value of previous food prices. The absence of an increase in food prices had no discernible impact on changes in food demand from the EU19.

The EU is dependent on imports of products from various countries, including Russia. Our study shows that EU sanctions against Russia have no impact on EU19 food imports. Hence, we did not perform an impact analysis (PSM) on this variable.

There are two important reasons SAN has no impact on EU19 food imports from Russia. The first reason is that the study runs until October 2022. Meanwhile, the implementation of sanctions began a few months earlier, so the impact has been minimal and poorly quantified. Second, the EU19 is indeed cautious when it comes to managing imports (Černý et al., 2021). So far, the imposition of EU sanctions against a country has hampered trade in almost all sectors, except food. For example, imports of vegetable and animal products, as well as vegetable fats and oils, animal and live products, beverages, and tobacco, from Iran have increased despite EU sanctions. This is largely due to the large Iranian diaspora in the EU, as well as the numerous Iranian grocery stores in EU28 countries (Ghodsi and Karamelikli, 2022).

Furthermore, the EU realizes that the food sector is very sensitive to import bans and embargoes. Food import barriers have the potential to cause food scarcity and price increases. Food import barriers can also stifle economic growth due to the EU19’s characteristics as an industrialized country. This area requires raw materials, including from the agricultural sector, to carry out its industrial activities. Hence, EU sanctions only apply to bilateral food trade between the EU and Russia; however, third-country individuals and companies can import food products from Russia if they are not on the EU sanctions list and do so entirely outside the EU (European Union, 2022).

Russia will also not remain silent in the face of EU sanctions, which could worsen its economic situation (Gurvich and Prilepskiy, 2015). However, the impact was not as severe as it was at the beginning and during the COVID-19 pandemic (Kuvalin et al., 2022). Russia has mandated that importers pay for trade transactions in rubles. Ruble transactions will significantly reduce the impact of Western financial sanctions (Timofeev, 2022). Russia is also attempting to diversify its product market beyond the EU (Nasir et al., 2022). China’s market is the most diverse and appealing to Russia (Timofeev, 2022). Russian companies are quickly adapting to sanctions by lowering investment and personnel costs, seeking new suppliers, and launching new products and modernization initiatives (Kuvalin et al., 2022).