The War in Ukraine and Its Impact on Global Agricultural Trade

The paper is structured as follows. Firstly, a literature review describes the general impact and economic consequences of the Russian full-scale invasion of Ukraine on the global economy. Secondly, it unveils why Ukraine turned out not to be ready for the military aggression of the Russian Federation in 2014 and was not able to prevent it or be maximally prepared for the aggression in 2022. Thirdly, the economic impact of the Russian invasion of Ukraine on the global trade of agricultural commodities is described by analysing global trade data. Finally, the findings are discussed in the light of actual and possible political and economic stimulus measures, and conclusions are drawn regarding how to ameliorate the impact of market disruptions on import countries of Ukrainian agricultural commodities.

The objective of the study is to assess the impact of the war in Ukraine on the global trade of agricultural products as a basis to design possible strategic measures to re-establish full participation of Ukraine in global trade. To achieve the stated purpose, the method of systematic analysis and comparison (to analyse the structural changes caused due to processes under study) is applied.

The methods used in the research are aimed at defining the degree of interconnection between the consequences of the war in Ukraine and the global trade and global GDP indices. To do so, Pearson correlation coefficient (cor) based on several significant suppositions has been calculated. Firstly, it is considered that the analysed variables have normal distribution and they have a linear interconnection. Secondly, should there be no connection between the variables when they vary independently, the value of the correlation coefficient equals null (r = 0). The stronger the connection between the variables, the closer the value of the correlation coefficient is to 1. Positive values of r indicate a positive (direct) connection, while negative values indicate a negative (reverse) connection. Along with the Pearson correlation coefficient, a number of degrees of freedom (df) and p-values were calculated in the research (Vitlinskyi et al., 2014).

To range factors of changes in the global trade, which were caused by the Russian aggression against Ukraine, a multiple regression model was built using the function lm() in programming environment R, which is used for statistical calculations. The multiple regression model enables defining dependence between the variables and forecasting the value of a dependent variable based on the independent factors. The formula of the multiple regression model is as follows: Y=β0+β1X1+β2X2+⋯βpXp+ε, {\rm{Y}} = {\beta _0} + {\beta _1}{{\rm{X}}_1} + {\beta _2}{{\rm{X}}_2} + \cdots {\beta _{\rm{p}}}{{\rm{X}}_{\rm{p}}} + \varepsilon, where –

Y is a dependent variable that we are attempting to predict;

The aim of the regression model is to estimate regression coefficients (β₀, β₁, β₂, …, β_p) and to use these coefficients to predict the values of the dependent variable Y based on the input values of the independent values (X₁, X₂, …, X_p). While building the model, the function step() was used to select the least significant factors that do not have considerable impact on the explanatory variable. This selection provides a more accurate and significant model. The modelling resulted in obtaining a number of coefficients, and based on their values, the value of the dependent variable can be forecasted. Major indices obtained from calculating function lm() are as follows:

Intercept (a point of intersection of lines with the coordinate axis or interception): This value indicates where the regression line crosses axis y if x = 0. It is calculated using the formula: Intercept=β0 {\rm{Intercept}} = {\beta _0}

Additional tests to check the model suppositions, including normality of distribution residuals (using the Shapiro–Wilk test) and autocorrelation (using the Durbin–Watson statistic), were carried out. The Shapiro–Wilk test is used to check the hypothesis regarding the normal distribution of model residuals. In this test, W is calculated based on the residuals order statistic and is compared with theoretical distribution, which is considered normal. The p-value of the test demonstrates how serious the deviation of the observed data is from the normal distribution. If the p-value exceeds the established level of significance, the hypothesis about normal distribution is not rejected. The Durbin–Watson statistic is used to check for autocorrelation in the model residuals. This statistic is based on comparing adjacent residuals, and it estimates the degree of lineal dependence between them. The value of the Durbin–Watson statistic varies between 0 and 4. A value close to 2 indicates the absence of autocorrelation, a value under 2 indicates positive autocorrelation and a value over 2 indicates negative autocorrelation. Both tests use input data (model residuals) to calculate statistics and to compare them with statistical distribution. Statistical values are compared with an established significance level, so as to draw a conclusion concerning distribution normality and autocorrelation in model residuals. Therefore, the built regression model and calculation of these indices let us estimate the impact of the factors on the variable and forecast values of the dependent variable based on this model.

Uncertainty about the duration and geographical scale of the war causes intensification of fluctuating changes within the global agricultural production system. Currently, several direct and indirect consequences of the Russian invasion of Ukraine may be defined. The direct consequences include the considerable reduction in production and, consequently, export of both countries, destruction of infrastructure, decrease of human resources caused by mobilisation of the population to the military, etc. Among the indirect consequences are the economic sanctions introduced by the international community towards the Russian Federation, the aggressor country. The sanctions aim to shorten the duration of the war, and even though the scope and reach of them is unprecedented since World War II, it is unclear if they will succeed to bring the conflicting parties to the negotiation table (Mulder, 2022). Sanctions have some flaws: first, the time lag between measures and visible impact on the Russian economy; second, the fact that embargos are not being enforced by over 100 countries representing 40% of world GDP; and third, the retaliation from Russia by threatening the supply of gas and oil to Europe. The European dependence on Russian fossil fuels is a painful reminder of the lack of strategic foresight of European economic policy, which some might see is characterised by geopolitical naivete (The Economist, 2022).

The above-mentioned events will result in a decrease in GDP within the countries taking part in the conflict as well as outside, particularly European countries, considering their close geographical proximity to the region and their dependence on Russian energy resources, which may destabilise their political systems. It is expected that the world GDP will grow by only 2.8% in 2022 in terms of market currency rates compared to the increase by 5.7% in 2021. However, long-term effects may be graver. In particular, there is a risk of further aggravation of fluctuation processes, limiting competition, and reduction of innovations within the global economic system, which may lead to a decrease in the world GDP by 5% in the long-term perspective (WTO, 2022b). In addition, according to experts' assessment, in 2022, the war will have greater impact on global trade rather than on the global GDP. To estimate the degree of influence of the war in Ukraine on major indices of the global economic system, a series of correlation coefficients was calculated (Table 1).

While carrying out the correlation analysis, the index of cereal production in Ukraine was taken as an independent variable since the country is one of the key exporters to the international agricultural market. As a result (Table 1), the correlation between the world GDP and international trade fluctuation (0.81) and production volumes of the mentioned produce (0.75) was calculated. In addition, greater manifestation of war consequences in Ukraine will be observed in global trade. Forecasts estimate a decrease in growth to 1% in 2023, down sharply from the previous estimate of 3.4% (WEF, 2022; WTO, 2022a, 2022c). It is important to note that the growth rate of global trade could be influenced by many factors, including economic conditions, political developments and changes in trade policies, among others. These figures may be reconsidered due to the indeterminacy of the conflict development in Ukraine. Therefore, on the global scale, a consequence of the Russian–Ukrainian War is a tendency of growth rates slowing down production processes as well as trade turnovers within the Ukrainian and international agricultural product sectors. Simultaneously, in the near term, trade costs will rise due to export limitations, logistics failures, the high cost of energy sources, etc.

As a result of these circumstances, agricultural product supply failures have caused a drastic price hike. According to FAO food price indices (FFPI), international export quotations of major foods have been growing since the second half of 2020 almost uninterruptedly. This ascending tendency climaxed in March 2022, when they reached their historical maximum (Fig. 1).

Figure 1.

At the same time, the global markets of cereals and vegetable oil are those that suffered most from the price hike (FAO, 2022c). Particularly, wheat cost has risen by over 40% since the war started (herewith, futures prices have increased by over 60%) (European Commission, 2022). Prices for foods such as rice and corn have acquired a similar tendency due to production hitches and because these goods are a wheat substitute. Cereal and vegetable oil prices continued growing during the first 5 months of 2022. As for wheat, prices grew in March due to shortened export stocks before the harvest of 2022–2023. The situation has been further deteriorating due to Ukrainian export failures because of the logistics crisis (FAO, 2022d). Global prices for different kinds of meat have similarly risen. Poultry meat has undergone the highest price rise, since its supply in the global market is limited too due to the war in Ukraine and bird flu outbreaks. Even though prices slightly decreased over the following 2 months, quotations of all product groups in FFPI remain high. Higher food prices mostly affect countries with low- or medium-income levels in Africa, southwards from the Sahara Desert (Botswana, Zimbabwe) and in the Near East (Algeria, Tunis), intensifying the poverty and food crisis issue in the world (WTO, 2022b).

Based on the multiple regression method, considerable dependence of global food prices on the volumes of production and export of major kinds of agricultural products from Ukraine was computed (see Table 2). The model uses several independent variables (provided in Table 2 in the indices column, except for dependent variables), which were used to calculate the degree of dependence of a dependent variable (food price index [normal] in one case and supply world cereal market [million t] in the other). To level the ability of determination coefficient R² to increase along with several variables in the regression model, the adjusted determination coefficient (adjusted R-squared) was calculated. As a result, food price index (normal) appeared to have a high degree of density of connection and supply world cereal market (million t) appeared to have a moderate degree of density of connection. According to the residual standard error (RSE) index, the model meets the data more accurately in the case of calculations for food price index (normal). A similar model was built for cereal supply within the global market. Its results enable us to state the presence of a direct connection with the last index.

Having used the mentioned data and the regression model in the first formula, we obtained the following outcomes: the value of multiple R-squared equals 0.9613 and the value of adjusted R-squared equals 0.9381. It means that the model used in the regression analysis accounts for 96.13% (or 93.81% considering the number of predictors) of variation in food price index (normal). RSE equals 3.212. It measures the volume of residual distribution around the regression line. The less the value of the standard error is, the better the model explains input data variation. The F-statistic equals 41.41 with 3 and 5 df. It is used to estimate the model significance on the whole. The p-value for the F-statistic equals 0.0005916, which is under the established significance level of 0.05. It means that there is statistically significant proof that at least one of the predictors is significant in the model. The outcomes of the Shapiro–Wilk test on residual distribution normality demonstrate that the W-statistic equals 0.955, and the p-value is 0.7449. The p-value exceeds the established significance level of 0.05, which means that there is no valid proof of rejecting the null hypothesis about normal residuals distribution. The Durbin–Watson statistic value equals 2.62. The value is close to 2, which demonstrates the absence of significant autocorrelation in model residuals. Summing up, the model has a high determination coefficient, which means that it perfectly explains the dependent variable. There is also no proof of deviation from residuals distribution normality and the absence of autocorrelation.

As for the second formula, the following outcomes were obtained: the multiple R-squared value equals 0.7173 and adjusted R-squared is 0.6231. This means that the model used in the regression analysis accounts for 71.73% (or 62.31% considering the number of predictors) of variations in supply world cereal market. RSE equals 98.75. It measures the volume of residual distribution around the regression line. The less the value of the standard error is, the better the model explains input data variation. The F-statistic equals 7.613, with 2 and 6 df. The p-value for the F-statistic equals 0.02258, which is under the established significance level of 0.05. This means that there is statistically significant proof that at least one of the predictors is significant in the model. The outcomes of the Shapiro–Wilk test on residual distribution normality demonstrate that the W-statistic equals 0.94661, and the p-value is 0.653. The p-value exceeds the established significance level of 0.05, which means that there is no valid proof of rejecting the null hypothesis about normal residuals distribution. The Durbin–Watson statistic value equals 1.92. The value is under 2, which may indicate the presence of positive autocorrelation in model residuals. Summing up, the model has a moderate determination coefficient, which means that it explains part of the dependent variable variation. Residuals distribution normality has not been rejected, but positive autocorrelation may be present.

It must be taken into account that the research used limited data of a particular time period. It might limit generalising outcomes in other time periods or in other contexts. The regression model involves some suppositions, such as lineal dependence between variables, residuals independence, and normal residuals distribution. These suppositions may not be fulfilled perfectly, which might affect the accuracy of the outcomes. Moreover, the model does not include all possible factors that may influence a dependent variable. Other factors not considered might have an impact on the relationships between variables and result in a change of outcomes. Despite the relatively small Ukrainian share in general volume of the global trade and production (Table 3), its integration into international agricultural markets is considerable. The Ukrainian agrarian sector is a powerful exporter of agricultural products with several items, including major foods. According to Ukrainian customs statistics, in 2020 compared to 2019, agrarian sector product export accounted for 22.9 billion USD, which is higher by 0.2%, while import accounted for 6.9 billion USD, which is higher by 13.2%. In other words, the surplus of foreign trade balance decreased by 4.3%. In 2020, the share of agricultural commodities in Ukrainian export of all products accounted for 47%, which is more by 0.9 p.p. compared to 2019 (UKRSTAT, 2022). In 2018–2020, Ukrainian export of sunflower seeds, corn, and wheat accounted for 38%, 10.6%, and 7.2%, respectively, within the global market (Ukraine took the first, fourth, and fifth place in the exporter rating, respectively). In 2021, the given tendencies strengthened (Table 4), and such a trend was supposed to remain.

In the 2020–2021 marketing year, Ukraine produced a third of the global sunflower oil, nearly half of its global export, and two-thirds of sunflower meal (Table 5). However, as of 2022, after the full-scale invasion of Ukraine began, the number of export and import deals considerably decreased compared to the similar time period in 2021 (UKRSTAT, 2022). Before 1 March 2022, Ukraine had dispatched around 61% of the forecasted export of sunflower oil of 2021–2022. As for rapeseed, soy seed, and corn, Ukraine is the sixth largest producer (the ninth for soy) and the third, seventh, and fourth largest exporter, respectively. As of March 1, Ukraine managed to dispatch about 65% of the forecasted soy export and 80% of corn from the general planned volume in the 2021–2022 marketing year. Before the Russian full-scale invasion of Ukraine, Ukrainian sunflower oil had taken the highest specific weight in the import cost of EU countries (24% of the general EU import from Ukraine), while corn (24%) and rape (15%) followed. As a quantitative indicator, the Ukrainian share in the EU import had reached 87%, 53%, and 35%, respectively. In March 2022, the import of these three products to EU decreased by 10% (sunflower oil), 37% (corn), and 29% (rape), respectively, compared to the same time period in the previous year. Ukraine is the seventh largest wheat producer in the world and, as it is predicted, would become the fifth largest exporter in the 2021–2022 marketing year (FAS, 2022b).

Generally, in 2020, major Ukrainian partners in agrarian and food product export were China (16% of the general export volume of Ukrainian agricultural products), India (6.7%), the Netherlands (6.4%), Egypt (6.2%), Turkey (4.8%) and Spain (4.4%). Concerning imports to Ukraine, the major partners were Poland (11.3% of the general import volume of Ukrainian agricultural products), Italy (7.3%), Turkey (7.3%), Germany (6.9%), the USA (4.4%) and the Netherlands (4.2%) (Table 5). At the same time, countries that are currently most vulnerable to the described crisis are those importing domestic agricultural products, like Gambia, Lebanon, the Republic of Moldova, Djibouty, Libya and Tunisia, whose general wheat import from Ukraine reached over 40% (Ivchenko et al., 2021). These countries have a low income level per capita, since their population usually spends a greater part of its income on food compared to developed countries. Because of the great dependence on the import of domestic products and considerable volatility of domestic markets, they may encounter challenges of rapid transition to alternative export sources, which will likely lead to a delivery deficit in the short-term perspective. Ukrainian wheat production and export hitches determine particularly high destabilising risks in this segment, considering the prewar production and trade volumes.

Countries that directly depend on food imports from Ukraine need to find other suppliers as soon as possible. In the long term, to guarantee food security to their population, such countries should pay attention to diversifying their imports and their domestic production. Moreover, the international community must pay attention to existing potential long-term challenges concerning global growth and cooperation and set measures to counteract the world being ‘separated’ by the aggressor country.

Therefore, the war in Ukraine has caused further destabilisation of the global agrarian markets in times of already highly volatile international product and raw material prices. Considering favourable weather conditions and some growth of production facilities of the Ukrainian agrarian sector, potentially, 2022 was supposed to deliver high yields of cereals, accordingly maintaining a high level of export opportunities. However, military actions in Ukraine and invasion of a considerable part of its territory provoked several negative factors, particularly the relocation of population, damage and destruction of civil infrastructure, limiting transportation of people and goods, which partially or fully froze the ability of farmers to act normally, harvest and sell their harvest. The situation has further deteriorated due to interruptions in infrastructure and institutional provisions. After the active military actions moved to the eastern and southern parts of the country at the beginning of April 2022, economic activity, particularly agricultural activity, in the central, northern and some southern parts of the country started to resume. In these areas, farmers started working in the fields, often facing a challenge to retrieve unexploded ammunition before applying fertilisers for winter crops or preparing soil for planting spring crops. These circumstances evoke serious anxiety about their potential negative effect on the food security within the country as well as at the international level.

For the international organisations to overcome damaging war consequences, countries' governments have developed and sent various kinds of aid to Ukraine. Particularly, substantial effort was made to handle logistical problems. Since the beginning of the full-scale invasion and blockade of Ukrainian ports, through which about 90% of crops were shipped abroad, the issue of their export possibilities was aggravated. To solve the problem, negotiations started and alternative logistic supply chains were created, particularly through the seaports of neighbouring countries (i.e. Bulgaria and Romania). Through the mediation of Turkey, an agreement to create a safe export corridor in the Black Sea water area was signed 22 July 2022. At the same time, alternatives were created for blocked cereals to be transported from Ukraine. For the same reason, according to the agreement, some sanctions concerning cereal export from Russia were weakened too and a new net of storage facilities for agricultural products in Ukraine was created. In addition, considerable support has been directed to agricultural producers to stimulate crop production (USAID, 2022). This so-called ‘grain agreement’ signed separately by Ukraine and Russia on one side and representatives of Turkey and the UN on the other side (22 July 2022) foresees delivery of 20 t of grain that were stuck in Ukrainian Black Sea ports (Anon, 2022). Such an agreement can have a significant impact on global food security and help restrain the food price levels. The inability of Ukraine to export grain from its Black Sea ports has drastically decreased food supplies in countries in Africa and the Near East, which greatly depend on such imports.

The above-mentioned agreement will allow to partly reduce the pressure on grain exports from Ukraine and will avoid shortage in storage facilities. According to the data of IGC, Odessa and Black Sea ports are capable of exporting from 5 to 6 million tonnes of grain monthly, should they work at full capacity. Considering the current export power of 2.2 million tonnes, creating opportunities for extra trade channels through Black Sea ports would be enough to satisfy the needs of 7 million tonnes of export power (IGC, 2023), which is necessary to prevent a shortage in storage capacities; however, missile strikes in Odessa and Black Sea ports that followed the signing of the agreement caused serious infrastructure losses and considerably decreased the export capacity of these ports.

It must be mentioned that apart from the described efforts of the international community, any assistance to increase the opportunities of freer export of domestic products without tariffs or non-tariff barriers to foreign markets is highly important for Ukrainian industrial agriculture in this complicated situation. In particular, Poland has accelerated undergoing customs procedures and Ukrainian product transportation, G7 countries fully support creating new selling chains by Ukraine, and Austria has lifted restrictions on Ukrainian carriers. Now, cargo traffic in Austria is available without any permits, which, along with previously lifted restrictions by Slovakia and Italy, enables Ukraine to deliver its agricultural products freely to Italy and its sea ports (Agravery, 2022). A number of international partners upgraded or developed programmes for supporting agrarian production in Ukraine. In particular, a plan for assessing the needs to reconstruct and restore Ukraine has been created by FAO (2022a) and the European Bank for Reconstruction and Development will facilitate credit resources to Ukrainian agrarian companies.

Along with support from abroad, which is already being given to Ukraine, the economic help provided by Ukraine itself is critically important. Particularly, the government lifted restrictions on volumes of state guarantees on a portfolio basis on 15 March 2022 (KMU, 2022a) and the state programme of soft loans was broadened (KMU, 2022b).

So far, the Ministry of Agrarian Policy of Ukraine has decided that promoting oil crop exports is its top priority, considering their higher value and the fact that they cover logistical expenditures. At the same time, corn and wheat are likely be stored in the country for 6–12 months (UGA, 2023). Considering this, the Committee on Finances, Tax and Customs Policy of Verkhovna Rada of Ukraine has recommended approving a bill on exemption from import duty taxation of goods used for storing grain and/or oil crops (# 7548-1) in the first reading of the bill and also generally (KMU, 2022c). It may be assumed that taking such measures will help Ukraine not to sell agricultural commodities at discount prices and save its products and functioning agrarian production in the mid-term perspective.

It must also be mentioned that the means for regional state administrations to grant subsidies have been increased. To create food security on the domestic market, the government of Ukraine implemented zero quotas on export of particular types of agricultural products on 5 March (Verkhovna Rada of Ukraine, 2022). Particularly, the export of living livestock, meat and meat by-products, buckwheat, wheat, sugar, barley, millet and some other items was banned. All these measures are aimed at protecting the majority of Ukrainian agrarians from defaults and to provide them with efficient volume of liquidity, whose full loss compensation is impossible in the short term. Many producers have currently fully halted their activity. However, the measures taken will help to provide food security in Ukraine and decrease the crisis effects for partnering countries.

Therefore, the Russian full-scale invasion of Ukraine led to several market disruptions for global trade of agricultural commodities and redistributed market power and supply infrastructure on the global agrifood market. In sum, this development weakened the position of countries of the Black Sea region that were leading exporters of agrarian products. Aiming at preventing famine, countries that participate in the global trade system have replaced tariff trade regulations with non-tariff measures to regulate trade volumes.