Implications of War in Ukraine for Polish Food Prices: Evidence and Mechanisms

Agri-food products are among the key goods satisfying basic human needs. One of the food security criteria is economic accessibility, which ensures that society as a whole or individuals have the necessary purchasing power to obtain an adequate amount of physically available food (Maciejewski, 2018). Among the factors that led to significant market disruptions and food security consequences was the war in Ukraine, which triggered a series of events with global consequences, especially in terms of energy and food supplies. Disruptions in these markets increased both global and local prices of agri-food goods (Arndt et al., 2023; Ben Hassen and El Bilali, 2022). Along with macroeconomic factors, this translated into food inflation growth in Europe (Kornher et al., 2024) and worldwide (Adjemian et al., 2024; Algieri et al., 2024; Hamulczuk et al., 2023b).

In recent years, Poland has also experienced inflation increases, both in general and food prices. The rise in food prices was already observed during the Covid-19 pandemic (Jędruchniewicz and Wielechowski, 2023); however, its peak occurred after the outbreak of the war in Ukraine. Although this war is widely indicated as a contributing factor to the surge in retail food prices in Poland, no in-depth studies on this subject have been found in the literature, indicating a significant research gap. Numerous studies (e.g., Budzyńska and Kowalczyk, 2024; Hamulczuk et al., 2023a) have demonstrated the war in Ukraine impacts on how the Polish agri-food sector functions and on selected commodity prices; however, there is no scientific study in the literature that quantitatively indicates the true rise in consumer food prices as a result of the conflict. Therefore, the primary objective of our paper is to examine the impact of the war in Ukraine on changes in Polish retail food prices. We sought to identify not only how much Polish food prices have risen due to the Russia-Ukraine conflict, but also to investigate the transmission mechanisms of this conflict for food prices in Poland. Understanding the determinants of prices allows consumers and producers to better comprehend the causes of price changes and make more informed decisions. At the same time, it is crucial for policymakers to make evidence-based decisions regarding monetary, fiscal, and trade policies. Therefore, the authors make a significant contribution to the existing body of knowledge.

In a globalized world, the impact of an armed conflict like that which erupted in Ukraine in 2022 cannot be confined to a single country or region. The war has significantly impacted global supply chains, particularly in the agri-food sector, where Ukraine is a key exporter of grain and oilseeds (Arndt et al., 2023; Ben Hassen and El Bilali, 2022). The blockade of Black Sea ports and traditional export routes significantly disrupted supply, forcing countries to seek alternative routes through EU countries. Initiatives such as the Black Sea Grain Initiative (BSGI) agreement aimed to mitigate these challenges, enabling grain exports to resume despite the conflict. However, border closures and heightened customs and logistics controls resulted in additional delays and higher transportation costs, which in turn drove up global commodity and food prices (Budzyńska and Kowalczyk, 2024; Hamulczuk et al., 2023a). In the initial phase of the conflict, food security became a significant challenge for many countries, particularly those heavily reliant on agricultural imports from the region.

The war’s negative consequences on agricultural and food prices are widely discussed in the literature. However, conclusions drawn by the authors depend on the time and a country/region perspective. At the beginning of the war, world agri-food prices rose sharply (Glauben et al., 2022; Nasir et al., 2022). However, in the long run, we observed a decline in world agricultural commodity prices due to global supply responses and the conclusion of the BSGI (Carter and Steinbach, 2024). The short-term increase in global agri-food prices is in line with simulations conducted immediately after the outbreak of the war by Berndt et al. (2022).

This dramatic surge in global prices in the first phase of the war stemmed from a physical reduction in supply, combined with growing uncertainty about future harvests and export opportunities (Ihle et al., 2022). However, the shock to the commodity market caused by the conflict affected not only the agri-food sector: Russia, Belarus, and Ukraine were also significant suppliers of coal, crude oil, natural gas, and mineral fertilizers (Alexander et al., 2022). Trade restrictions and embargoes imposed on the Russian Federation led to significant worldwide price rises in the energy market. As indicated by Just and Echaust (2023), a significant increase in volatility spillover between energy and grain prices occurred after an outbreak of war. The war-induced surge in energy and fuel prices affected the international agri-food market, translating into higher agricultural production costs, processing costs, and transportation costs.

Furthermore, high energy and fuel prices weakened household purchasing power, reducing families’ ability to purchase food (Arndt et al., 2023). The problem of rising retail food prices did not only concern low-income and food-import-dependent countries. Kornher et al. (2024) emphasized the negative impact of the conflict on inflation processes in the EU countries, mainly due to increases in energy prices and transportation costs. Rising food production and storage costs are primarily reflected in retail food prices in regions where food production processes are multi-stage. Developed countries consume more processed food, making the production and processing of food products more complex and energy-intensive. Consequently, the final product price largely depends on changes in marketing costs (Dębkowska et al., 2023; Hamulczuk et al., 2023b). Apart from supply-side factors, Adjemian et al. (2024) stress that the money supply is a significant driver of food inflation in the US. This issue was also emphasized by Jędruchniewicz and Wielechowski (2023), who noted the lagged impact of the increase in the money supply on fluctuations in agricultural prices in Poland.

Poland, also an EU member state, likewise experienced rising retail food prices during the war. However, few researchers have addressed this issue. The war-induced increase in fuel and energy prices in turn contributed to the rising indirect costs of food production and transportation (Dębkowska et al., 2023). Furthermore, the decline in Ukrainian workers in Poland has intensified labour shortages, driving up labour costs. As a consequence, the combined disruptions in production, trade, transportation, and labour availability could have significantly impacted the overall increase in food prices felt by consumers in Poland (Krzykowski, 2022). Supply chain disruptions and higher operating costs resulting from the war also proved significant, as noted by Zając and Bogusz (2024). It is worth emphasizing that higher food prices in Poland occurred despite strong supply-side pressure in Ukrainian agricultural goods.

Our empirical research focuses on the short-term consequences of Russia’s aggression against Ukraine on retail food prices in Poland. According to Eurostat data, food price analyses were based on the Harmonised Indices of Consumer Prices (HICP) from Eurostat. The monthly food and non-alcoholic beverage price index (2015 = 100) and its main components (10 aggregates) from January 2010 to December 2023 were used. To estimate the impact of the war in Ukraine on changes in food prices in Poland, we compared the actual level of price indices with the forecasts of these variables obtained from forecasting models estimated on the basis of data from before the outbreak of war (till February 2022). Positive error values indicate that the conflict raised prices, whereas negative errors indicate that food prices decreased. A similar approach was used by Hamulczuk et al. (2023b) in assessing the impact of the war in Ukraine on global food prices.

The percentage forecast errors calculated for individual price aggregates were the starting point for further qualitative analysis. To explain the obtained results and facilitate scientific discussion, we referenced other data and information, including the export of agricultural goods from Ukraine (Comtrade, Eurostat-Commext), world food prices (OECD-FAO), and food prices in other EU countries (Eurostat).

Ex-post forecasts of retail prices (price indices) were calculated using univariate regARIMA models, which means that the model learns from the data and makes predictions based on patterns (trends, seasonality) that it identifies in the past. Its merits lie in its having minor requirements for data, coupled with relatively high prediction accuracy. The disadvantage is that it does not delve into the economic mechanism of data generation. It is therefore important to evaluate the obtained results and considering other information qualitatively. Comparing actual data with counterfactual values derived from forecasting models can be an alternative to approaches based on simulations or econometric models. The approach used is undoubtedly easier to implement due to the relative simplicity of these models and the low data requirements. In other alternative simulation models, a no-war scenario would also need to be estimated.

RegARIMA models were estimated using a price series from January 2010 to February 2022, data covering the pre-war period. In contrast, ex-post forecasts were calculated for the war period from March 2022 to December 2023. The regARIMA is a part of the preliminary modelling within the well-known and widely used X-13 ARIMA-SEATS deseasonalization procedure (X-13..., 2020). RegARIMA model has the following form: (1) Yt=∑inβiXi,t+Zt {Y_t} = \sum\nolimits_i^n {{\beta _i}{X_{i,t}}} + {Z_t} where: Y_t – original time series, β_i – parameter at the i-th explanatory variable, X_it – i-th explanatory variable, Z_t – residual from the model, estimated using the SARIMA(p,d,q)(P,D,Q)s model (Box and Jenkins, 1970): (2) φBΦBS1−Bd1−BSDZt=θBΘBSεt \varphi \left( B \right)\Phi \left( {{B^S}} \right){\left( {1 - B} \right)^d}{\left( {1 - {B^S}} \right)^D}{Z_t} = \theta \left( B \right)\Theta \left( {{B^S}} \right){\varepsilon _t}

In SARIMA model d and D are the orders of non-seasonal and seasonal differentiation, p and P are the numbers of non-seasonal and seasonal autoregressive lags, q and Q are the numbers of non-seasonal and seasonal lags in the moving average, s is the number of seasons in a year, φ, Φ, θ and Θ are the model parameters, B is the backward shift operator and (1 – B) is the differentiation operator. Among regARIMA regressors, we can include variables responsible for: additive outlier AO, level shift LS, temporal shift TS or Easter effect.

We can see that the regARIMA model is an extension of the SARIMA model. Apart from autoregressive and moving average lags, it includes additional explanatory variables to cover the impact of moving holidays, outliers, or structural changes in the time series. The estimation of regARIMA models was preceded by a unit root test, which revealed that all series were I(1). The specification of the regARIMA model covers selecting the regression variables and their timing, and the number of ordinary and seasonal differences, as well as the number of autoregressive and moving average lags. It was performed within the X-13-ARIMA procedure using information criteria. For more information on the regARIMA or X-13-ARIMA models, see, among others, the X-13 ARIMA-SEATS Reference Manual (2025).

To test the robustness of the research results, analogous ex-post forecasts were estimated using the multiplicative exponential Holt-Winters model (Hyndman et al., 2008). In this model, the forecast is a function of three components: level, trend, and seasonality, whose levels depend on ‘smoothing constants’. It is recommended for irregular (stochastic) changes in phenomena. Food prices belong to this category of economic phenomena. More on this model can be found in Hyndman et al. (2008).

Generally speaking, both models are based on the assumption that existing price trends and patterns will continue (see Fig. 1 as an example). Therefore, if a new event occurs–in our case, a war–that changes price dynamics, the differences between prices and their forecasts can be attributed to it. Their main drawback is that the war is assumed to be the only factor influencing forecast errors in our case. However, it appears that during this period, the war in Ukraine and all economic activities were primarily focused on mitigating the negative effects of the conflict. Other factors played a lesser role, although their influence cannot be ruled out. For example, the direct effects of the COVID-19 pandemic, related to restrictions, were short-lived, and the restrictions themselves did not play a significant role during the war, as Poland and most EU countries lifted most of them in early 2022. However, the pandemic undoubtedly led to a shift in monetary and fiscal policy, resulting in an increase in inflation. However, the rise in energy prices due to the pandemic was largely reflected in prices (and their trends) until the outbreak of the war, and such series were used to prepare ex-post forecasts for the war period. Therefore, while the effects of the pandemic on forecasting bias may appear minimal, they cannot be ruled out (second-round effects). Among the other exogenous factors, changes in sectoral conditions may have played a greater role, potentially affecting price dynamics only within a given commodity group.

Fig. 1.

Sample of the ex-post forecasting process and calculation of the ex-post errors for food prices

Source: own calculations based on RegARIMA models and Eurostat data.

This paper aimed to analyse the impact of the war in Ukraine on retail food prices in Poland, with a particular emphasis on the transmission mechanisms of external shocks to the domestic agri-food sector. Using a counterfactual approach, we compared expected (no-war scenario) price developments with actual data to assess the conflict’s impact on food prices. The general findings indicate that the war in Ukraine contributed to a sharp increase in prices of agricultural commodities, energy, fuels, and labour. This led to a significant rise in food prices in Poland. Kornher et al. (2024) confirmed that global market turbulence associated with the Russia–Ukraine war contributed to rising food price inflation in the EU countries. Retail food price increases in Poland affected all major commodity groups, with the most significant changes being observed in meat and dairy products between 2022 and 2023. The increase in fruit and vegetable prices was least noticeable after the outbreak of the war, which aligns with the findings of Kornher et al. (2024).

The mechanism by which the war impacted food prices in Poland is quite complex. Initially, the outbreak of the war in Ukraine led to restrictions on food exports, resulting in a short-term global rise in agricultural commodity prices. Given the relatively small share of agricultural raw materials in consumer prices, changes in the non-raw material costs of food production and distribution played a greater role in food inflation. Rising fuel and energy prices and increasing labour costs played a key role in the growth of food prices in Poland. The increase in fuel prices raised operational costs across the entire supply chain, while higher electricity and gas prices significantly impacted energy-intensive processing sectors. Maintaining low interest rates in the early phase of the conflict, along with numerous government support programmes, intensified inflationary and wage pressures. Adjemian et al. (2024) also attribute a large proportion of the food price inflation experienced in the US to the money supply. Therefore, the main recommendation is to adopt a more responsible monetary and fiscal policy in response to global inflationary processes. This recommendation is supported by the fact that during the period covered by the research, food inflation in Poland was among the highest in the EU, which raises questions about the effectiveness of the economic policy. It appears that the government and the National Bank of Poland failed to act rationally, or in a manner that prioritized non-economic criteria, by simultaneously maintaining strong support for households, low interest rates, and high energy prices for businesses. This fed into higher non-agricultural costs of food production, which were later passed on to retail prices.

Although the quantitative study results are quite robust when using an alternative impact analysis method, they have certain limitations. A key limitation of the analysis performed is the assumption that the war in Ukraine and its sectoral and macroeconomic implications were the primary drivers of food price increases in Poland during the research period. Therefore, future research should focus on more complex simulation models that would control for other factors, such as economic policies, interventions or agrometeorological conditions. Potential models may include econometric difference-in-difference models or the dynamic stochastic general equilibrium models.