Drivers of Global Wheat and Corn Price Dynamics: Implications for Sustainable Food Systems

Abstract

Globalisation, population growth, climate change, and energy-policy shifts have deepened interdependence between agri-food and energy systems, amplifying price volatility. This study examines the determinants of global wheat and corn price dynamics over 2000–2023, emphasising energy markets (oil and biofuels), agronomic and climatic factors, population pressure, and cross-market interdependencies. Using multiple linear regression with backward selection on annual global data from official sources (FAO, USDA, EIA and market series), we quantify the relative contributions of these drivers. The models explain most of the variation in world prices (R² = 0.89 for wheat; 0.92 for corn). Oil prices are a dominant covariate: a 1 USD/barrel increase in Brent is associated with a 1.33 USD/t rise in the wheat price, while a 1 USD/t increase in the corn price raises the wheat price by 0.54 USD/t. Lower biodiesel output per million people is linked to higher wheat prices (+0.67 USD/t), underscoring the role of biofuel supply conditions. We also document an asymmetric yield effect—higher yields correlate positively with wheat prices but negatively with corn—consistent with crop-specific market mechanisms. Although temperature and precipitation were excluded from the regressions due to collinearity, their strong correlations with yields and biofuel activity signal continuing climate risk. The contribution of this study lies in integrating energy, climate, and agricultural market factors within a single empirical framework, offering evidence of their joint role in shaping staple grain prices. These findings add to the literature on food–energy linkages and provide insights for sustainability policies, particularly the design of integrated energy–agriculture strategies and risk-management instruments to enhance resilience in global food systems.

1. Introduction

Population growth, climate change, armed conflicts, and new technologies in production and distribution are reshaping global agri-food systems and increasing price volatility. FAO (2023) reports that global food systems are vulnerable to weather, political, and market shocks that destabilise the supply–demand balance and threaten food security [1]. From 1960 to 2008, agricultural prices were relatively stable; since then, prices of staples such as rice, wheat, and corn have surged, heightening food-security risks [2]. Rising demand—particularly in populous economies such as India and China—and the expansion of biofuel sectors have compounded these risks [3]. Globalisation, technological change, and dietary shifts further amplify volatility [4]. Because agriculture depends on living organisms, it is uniquely exposed to climatic and biological shocks [5]. While prices reflect supply–demand fundamentals and policy settings, volatility is increasingly shaped by these external, systemic drivers, with clear implications for resilient and sustainable food systems.

Theoretical and empirical research has documented price linkages and their transmission along the vertical marketing chain [6,7,8]. More recent work using threshold cointegration [9] and asymmetric cointegration [10,11] offers a more nuanced view of short-run price responses to shocks and long-run adjustment. Further studies argue that cointegration is not a prerequisite for market integration, as transaction costs and other determinants of price spreads may evolve in complex, time-varying ways [12]. Collectively, this literature suggests that contemporary price volatility arises from non-linear and multi-layered transmission mechanisms.

Price co-movement across markets is pronounced, even in the absence of physical trade, driven by shared information flows [13]. Long-run transmission is observed among grains [14] and is exacerbated by globalisation, which fuels uncertainty and non-fundamental market behaviours [15]. Such volatility hinders sustainable production and progress towards Sustainable Development Goal 12 (Responsible Consumption and Production) [16].

Numerous studies have analysed price determinants in wheat and corn markets. Strong energy–agriculture linkages and climatic shocks are well documented [17]. Price volatility is shaped by oil prices, weather patterns, stock levels, and policy interventions [18,19,20] (p. 35). Recent research further highlights the roles of government policy and global disruptions [21].

Against this backdrop, a clear knowledge gap remains in identifying the key determinants driving price fluctuations in global wheat and corn markets. We still lack quantitative evidence on how energy markets, particularly oil and biofuel dynamics, interact with crop yields to shape wheat and corn prices over the long run. This study fills that gap by disentangling these interdependencies with a 23-year global dataset, offering new insights for stabilising grain markets and guiding sustainability policies.

This article aims to identify and assess the key factors influencing price volatility in the global wheat and corn markets over 2000–2023, with particular emphasis on the roles of energy markets (including biofuels), agronomic and climatic variables, population growth, and cross-market interdependencies. The analysis seeks to determine the principal drivers of volatility within an increasingly integrated landscape linking agricultural and non-agricultural markets and evolving global conditions.

We address this objective using multiple linear regression (MLR), enabling the identification of causal relationships among observed phenomena and the quantification of associations between the dependent variables—the world prices of wheat and corn—and a set of independent variables. These include average annual global prices of crude oil, biodiesel, and bioethanol; per-capita production of wheat and corn; and bioethanol and biogasoline output both per capita and per hectare of corn cultivation.

2. Determinants of Global Price Fluctuations in Wheat and Corn Markets

Heightened volatility in agricultural prices has significant implications for stakeholders across the food value chain. A substantial body of research has identified the principal drivers of price fluctuations in agricultural markets and mapped the network of interrelated factors that contribute to persistent instability [18,22,23,24,25,26].

To systematise these determinants, we classify them into three categories: supply-side factors, demand-side factors, and cross-cutting systemic drivers (

Table 1. The main factors of price fluctuations in agricultural markets.

Factors	Sources of Price Fluctuations in Agricultural Markets
Supply factors	Availability of arable land for agricultural production, including its alternative use (biofuel production)
	Degree of technical progress in agriculture
	Change in weather and climate conditions
	State of agricultural product stocks in the world
	Prices of production factors, including crude oil and gas
	Agricultural production seasonality
	Supply chain disruptions
Demand factors	Population
	Level of economic development
	Change in the structure of consumption
	Speculation on commodity markets
Other factors	Economic (state) interventionism
	Globalisation and changes in the macroeconomic environment
	Cyclicality of global food crises
	Pandemics and epidemics
	Warfare and geopolitical crises

Source: Authors’ synthesis based on [18,22,23,24,25,26].

).

Temperature and precipitation are primary determinants of cereal yields. The increasing frequency of extreme rainfall and heatwaves, together with floods and droughts, is inducing fluctuations in grain output and, by extension, pronounced movements in grain prices [27]. In 2024, global average annual precipitation was 1.29% below the 1980–2024 average, while the global average annual temperature was 1.61 °C higher than in 1980 (Figure 1).

Over the next five years, the fitted trend indicates a very high likelihood of global temperature shocks (R² = 0.8642). By contrast, the linear trend in average annual global atmospheric precipitation shows exceptionally low explanatory power (R² = 0.0347), which does not preclude numerous floods in some regions while, simultaneously, hydrological drought emerges in others. Results from the IPCC project, involving more than 300 climatologists, indicate that global temperatures will rise by at least 2.5 °C above pre-industrial levels by the end of this century [29]. This projected climate breakdown is primarily driven by rising fossil-fuel consumption and greenhouse-gas emissions. Growing food insecurity is well-documented and likely to increase global hunger [30]. Climate change has already triggered more frequent extreme weather events, reducing crop yields. Price impacts are partially offset by more resilient hybrids, improved storage technologies, regional and international trade dynamics, and risk-management instruments [27].

Population size and per capita production of wheat and corn determine effective demand for these cereals. The world population increased by 30.8% (approximately 1.9 billion people) between 2000 and 2024. In line with the trend, the world population is projected to rise by a further 7.4% in the next five years to 8.7 billion people (R² = 0.9995) (Figure 2). During 2000–2024, wheat production per capita increased only slightly (by 2.6%), suggesting that scientific, technical, and social developments did not materially alter traditional uses (consumption, processing, and animal feed). The linear trend in per capita wheat production is weak (R² = 0.2248). According to our calculations, per capita wheat production will remain in the range of 94–102 kg/capita until 2030. To secure wheat supply for the growing population, yields must increase by 11.4% to 3.9 t/ha by 2030 (with average wheat production of 97.8 kg/capita and an average wheat area of 219.2 million ha).

Per capita corn production increased by 50.6% over the analysed period, reaching 142.0 kg/capita, and the fitted trend indicates further growth of about 30.2% by 2030 to almost 185.0 kg/capita (R² = 0.887). Rising global energy demand—driven by rapid population growth, scientific and technical advances, and increasing societal demand for environmental protection—has reshaped the use of plant biomass, including corn, for energy purposes. To meet the growing demand for corn, its yields must increase by 31.6% to 7.5 t/ha by 2030 (assuming an annual increase in corn production of 8.4 kg/capita to 185 kg/capita and an average corn sown area of 202.8 million ha).

Meeting future cereal demand will also require wheat-production strategies that enhance physiological traits to raise yields irrespective of climate change. Multi-model projections estimate potential global wheat output at 1050 ± 145 million tonnes with improved photosynthetic performance, implying a 37% increase without expanding cultivated area [32]. The International Maize and Wheat Improvement Center (CIMMYT) identifies eight global mega corn environments, with yields ranging from about 6 t/ha in dry lowland tropics (e.g., East Africa, Central America, India) to roughly 14 t/ha in humid temperate regions (e.g., the US Corn Belt, Western Europe, Argentina) [33]. Yield variability in corn and wheat is shaped by genetic improvements, fertilisation, irrigation, and climate change [34]. Differences across core production regions indicate scope for intensification without expanding cropland, albeit likely at higher production costs, which could place upward pressure on cereal prices.

As noted above (Table 1), crude oil prices play a significant role in the formation of global wheat and corn prices (Figure 3).

The review period is characterised by pronounced co-movement in the prices of all three commodities, differing mainly in the amplitude of change. By 2024, wheat and corn prices were higher by 10.7% and 16.7%, respectively, relative to 2019, with fluctuations over 2019–2024 ranging from −20% to +115%. Brent crude prices rose by 31.9% in the same period, with swings from −60% to +100% (Figure 3). Pairwise correlations are high: corn–wheat (r = 0.8802), corn–crude oil (r = 0.9058), and wheat–crude oil (r = 0.9220). These results indicate strong co-movement and bidirectional volatility transmission between the cereal markets, while oil-market shocks exert a significant and persistent influence on wheat and corn price volatility.

3. Materials and Methods

Wheat and corn are the world’s principal cereals, cultivated across similar climatic zones and requiring comparable natural conditions and soil quality. Used globally for food and animal feed, they are exposed to common drivers of price formation and, consequently, often respond similarly to market fluctuations [38,39,40].

To assess the scale of influence exerted by key determinants on global wheat and corn prices, we estimate multiple linear regression (MLR) models. This approach relates a continuous dependent variable—the world price of wheat (and, separately, corn)—to a set of independent variables, whether categorical (after appropriate encoding) or continuous. The relationship is specified in the familiar linear form, where the dependent variable is expressed as a weighted sum of the predictors plus an error term capturing unobserved influences [41] (pp. 336–345):

$$Y_i = {\beta }_0 + {\beta }_{1 }{X}_{i1} + {\beta }_2{X}_{i2} + . . . + {\beta }_n{X}_{in} + e_i,$$

(1)

where $Y_i$ is the $i^{th}$ observation of the dependent variable for each observation i = 1, …, n; $X_{ij}$ of the $j^{th}$ independent variable, j = 1, 2, …, n. The values ${\beta }_j$ represent parameters to be estimated, and $e_i$ is the $i^{th}$ independent identically distributed normal error.

In the first step, we specified a baseline model with the full set of pre-selected variables. We then applied backward selection, iteratively removing the variable with the highest p-value at each step. The procedure continued until all remaining regressors were significant at p < 0.05. The choice of the backward elimination procedure was guided by both the structure of the dataset and the objectives of this study. Unlike penalisation techniques such as LASSO, which are more suitable for high-dimensional datasets with many predictors and potential sparsity, our analysis involved a relatively moderate number of theoretically pre-selected explanatory variables, grounded in prior literature and economic reasoning. Backward selection allowed us to retain model interpretability while systematically excluding variables that did not demonstrate statistical significance, thus avoiding overfitting and preserving the most parsimonious specification. Compared with stepwise selection, backward elimination reduces the risk of prematurely excluding variables that may later prove significant when considered jointly with others. Moreover, this approach is widely applied in econometric studies where theoretical considerations pre-define the pool of candidate regressors, ensuring transparency and alignment with established methodological standards in agricultural economics and commodity price modelling [42,43].

The econometric procedure incorporated standard diagnostic tests to mitigate the risk of spurious regression results. Prior to model estimation, all variables were tested for stationarity using the Augmented Dickey–Fuller (ADF) test. The results confirmed that the variables included in the regressions were either stationary in levels or had been appropriately transformed to achieve stationarity. To address potential autocorrelation in the residuals, the Durbin–Watson (DW) test was applied. The DW statistics for both models fell within the acceptable range, indicating no evidence of first-order serial correlation. Multicollinearity was further assessed through correlation matrices and variance inflation factors (VIF), and redundant variables were excluded from the final specification. This procedure minimised the risk of spurious relationships and enhanced the reliability of the estimated coefficients [41]. Overall, the modelling strategy—combining unit root testing, autocorrelation diagnostics, and multicollinearity checks—ensured that the statistical significance of the coefficients reflects genuine economic relationships rather than artefacts of trending time series.

The quantitative analysis draws on official data from international organisations (USDA; FAO) and from market and energy sources, including ICE Futures Europe and the US Energy Information Administration [28,37,44,45,46].

To build two MLR models, the average annual prices of wheat (${PR}_w$, grade 1, Rouen, USD/t) and corn (${PR}_c$, US No. 2, Yellow, USD/t) in the period 2000–2023 were selected as explanatory factors, according to FAO data [44].

The next step was to identify the factors explaining price volatility in wheat and corn markets (

Table 2. Selected factors influencing price volatility in the wheat and corn markets.

Variables	Units	Variable Descriptions
$P_{co}$	USD/barrel	Average annual crude oil prices according to data from the U.S. Energy Information Administration [46]
${GL}_p$	mm	Global land precipitation [28]
$T_a$	°C	Average temperature anomalies [28]
$A_w$	Wheat crop area, million ha	The area of cereal crops as a factor directly determining the level of supply of the examined cereals
$A_c$	Corn crop area, million ha
$Y_w$	t/ha (wheat yield)	Productivity of crops from 1 ha (determines the influence of weather and climatic conditions, which to the greatest extent determine the level of crop yields)
$Y_c$	t/ha (corn yield)
$P_{w}per$	kg/capita (wheat production)	Cereals’ production per capita (determines population growth, the level of nutrition of the global population and the elasticity of demand)
$P_c$per	kg/capita (corn production)
$P_{bf}$per	Barrels/1 million people	Biofuel production per 1 million people (determines the level of growth in demand and consumption of biofuels)
$P_{bg}{A}_c$	Barrels/1 thousand ha of corn crop area	Biogasoline production per 1 thousand ha of corn crops (determines the level of growth in the use of cereals and agricultural land for technical purposes)

Source: Authors’ study.

). These factors were derived from statistical information provided by official international sources of FAO and USDA, as well as supplementary datasets from the U.S. Energy Information Administration, Worldometers website, and the National Centers for Environmental Information [28,31,37,44,45,46,47]. The dataset used in this study covers the period of 2000–2023 (

Table 3. Selected factors explaining price dynamics in the wheat and corn markets, 2000–2023.

Years	${\mathit{P}\mathit{R}}_{\mathit{w}}$	${\mathit{P}\mathit{R}}_{\mathit{c}}$	${\mathit{P}}_{\mathit{c}\mathit{o}}$	${\mathit{A}}_{\mathit{w}}$	${\mathit{A}}_{\mathit{c}}$	${\mathit{Y}}_{\mathit{w}}$	${\mathit{Y}}_{\mathit{c}}$	${\mathit{G}\mathit{L}}_{\mathit{p}}$	${\mathit{T}}_{\mathit{a}}$	${\mathit{P}}_{\mathit{w}}\mathit{p}\mathit{e}\mathit{r}$	${\mathit{P}}_{\mathit{c}}$per	${\mathit{P}}_{\mathit{b}\mathit{f}}$per	${\mathit{P}}_{\mathit{b}\mathit{g}}{\mathit{A}}_{\mathit{c}}$
2000	147.9	88.3	26.7	215.1	136.9	2.7	4.32	818.7	0.59	95.6	96.36	29.4	1.23
2001	151.6	89.7	21.9	214.6	137.4	2.7	4.48	790.1	0.77	94.5	98.80	31.7	1.32
2002	175.9	99.4	22.5	214.9	137.5	2.8	4.39	772.9	0.94	93.9	95.68	36.8	1.54
2003	186.8	105.2	27.5	207.4	144.6	2.7	4.46	776.3	0.91	86.1	100.96	45.0	1.82
2004	186.1	111.9	36.9	215.7	147.5	2.9	4.94	794.0	0.72	98.1	112.73	50.0	1.95
2005	197.5	98.4	50.5	221.7	148.2	2.8	4.82	782.0	1.15	95.7	108.98	58.3	2.19
2006	216.8	121.4	59.6	212.6	148.2	2.9	4.78	812.4	1.03	92.6	106.69	73.3	2.65
2007	319.8	163.0	66.6	215.5	159.3	2.8	4.98	803.8	1.19	90.3	118.12	98.4	3.20
2008	299.4	222.2	94.2	222.1	163.7	3.1	5.07	815.4	0.89	100.0	122.01	131.5	4.11
2009	189.8	165.6	56.3	225.2	159.4	3.0	5.15	788.0	0.99	99.3	119.20	143.7	4.58
2010	230.8	185.3	74.6	215.6	165.3	3.0	5.16	823.2	1.19	91.9	122.34	163.4	5.07
2011	314.6	292.0	95.7	220.3	172.8	3.2	5.14	816.8	1.06	98.8	125.84	171.6	4.81
2012	310.6	298.3	94.6	217.8	180.4	3.1	4.85	790.7	1.05	94.3	122.48	172.0	4.54
2013	293.5	259.8	96.0	218.4	187.5	3.3	5.42	804.6	1.05	98.2	140.64	185.6	4.69
2014	255.5	192.9	87.7	219.5	186.5	3.3	5.58	774.8	1.09	99.6	142.19	200.0	5.01
2015	199.1	170.1	44.3	223.0	191.1	3.3	5.52	754.7	1.34	100.2	142.33	197.9	5.10
2016	179.1	159.3	38.4	219.0	194.1	3.4	5.79	787.5	1.64	99.9	150.02	202.4	5.00
2017	189.9	154.4	47.5	218.3	198.6	3.5	5.74	808.5	1.48	102.0	150.52	206.5	5.02
2018	220.4	164.5	61.5	214.0	195.3	3.4	5.76	788.2	1.35	95.6	146.78	228.2	5.47
2019	208.9	170.1	55.6	215.7	193.7	3.5	5.87	772.0	1.53	98.7	146.96	238.2	5.61
2020	227.9	165.4	36.4	217.9	199.3	3.5	5.80	782.1	1.67	96.8	147.79	221.4	4.92
2021	293.2	259.3	65.7	220.4	205.7	3.5	5.50	781.0	1.39	98.0	143.49	230.1	4.94
2022	366.3	317.7	94.1	219.2	203.5	3.6	5.98	788.5	1.41	98.2	153.15	239.8	5.16
2023	269.7	251.3	76.1	227.0	206.2	3.5	5.62	748.9	1.81	98.1	144.16	257.5	5.39

Source: Authors’ calculations based on NCEI (2024), FAO (2024), and USDA (2024) data [28,44,45].

).

Between 2000 and 2023, wheat, corn, and crude oil markets underwent marked shifts. Prices rose substantially—wheat by 1.8-fold, corn by 2.8-fold, and crude oil by 2.9-fold. Global annual wheat output increased by 33.4% (+197.3 million tonnes) and corn by 107.4% (+635.8 million tonnes). Relative to 2000, wheat productivity per hectare in 2023 was 26.9% higher, contributing 165.7 million tonnes (84.3%) to production growth, while wheat-sown area expanded by 5.5% (+11.9 million ha), adding 31.6 million tonnes (15.7%). For corn, productivity rose by 30.0%, lifting output by 177.4 million tonnes (27.9%), and sown area increased by 59.6%, adding 458.4 million tonnes (72.1%). Advances in production technologies supported productivity gains, while the expansion of corn area was largely associated with technological progress in biofuel production. From 2000 to 2023, biodiesel production per million people increased 8.8-fold, and bioethanol output per 1000 ha of corn sown rose 4.4-fold.

The next stage of the research tested the hypothesis that selected direct factors (sown area, yield, and production volume) and indirect factors (crude-oil price, global biofuel production and prices, humidity and temperature anomalies) exert a significant influence on price formation in global wheat and corn markets. Correlation and regression analyses indicated that only a subset of the initial variables was suitable for constructing the MLR models (

Table 4. Dependent and independent variables in the models.

Dependent Variables	Independent Variables
Model I—Price of 1 tonne of wheat (grade 1, Rouen), USD/t ( ${PR}_w$)	$P_{co}$—US Crude Oil First Purchase Price, USD/barrel; ${PR}_c$—the price of 1 tonne of corn (US No. 2, Yellow), USD/t; $Y_w$—wheat yields, kg/ha *; $P_{w}per$—wheat production, kg/capita; $P_{bf}$per—global biofuel production (originally published in barrels of oil equivalent per day), which for the purpose of this study is presented in barrels per 1 million people.
Model II—Price of 1 tonne of corn (US No. 2, Yellow), USD/t (${PR}_c$)	$P_{co}$—US Crude Oil First Purchase Price, USD/barrel; $P{R}_w$—the price of 1 tonne of wheat (grade 1, Rouen), USD/t; $Y_c$—corn yields, kg/ha *; $P_{bf}$per—global biofuel production (originally published in barrels of oil equivalent per day), which for the purpose of this study is presented in barrels per 1 million people.

* To ensure comparability across datasets, the yield index was converted to kilograms per hectare (kg/ha). Source: Authors’ study.

).

Factors $A_w$ and $A_c$ were eliminated from the model due to strong linear relationships with explanatory factors ${\mathrm{Y}}_{\mathrm{w}}$ and ${\mathrm{Y}}_{\mathrm{c}}$, indicating perfect multicollinearity. Factor $P_c$per and factor $P_{bg}{A}_c$ were also excluded from the model due to exact multicollinearity. The exact multicollinearity of factor $P_c$per resulted from the elasticity of demand for corn, i.e., the response of demand or quantity of corn offered on world markets to the change in its price, which was stronger than the price change itself. This explains the high demand for corn as a raw material for producing plant biomass for energy purposes. The exact multicollinearity of factor $P_{bg}{A}_c$ results from the linear relationship with another explanatory factor $P_{bf}$per [48] (p. 183).

The Global land precipitation (${GL}_p$) and average temperature anomalies $\left(T_a\right)$ were also excluded. Despite high correlation coefficients, their inclusion worsened regression statistics (t-values, p-values), undermining model credibility. Notably, $\left(T_a\right)$ correlated strongly with yields of wheat (79.9%), corn (83.0%), and biodiesel output per million people (82.2%).

Scatterplots (Appendix A, Figure A1 and Figure A2) confirm relationships between explanatory and explained variables. These plots illustrate correlation strength, shape, and direction, supporting the hypothesis of factor influence on global wheat and corn price formation.

In Model 1, the explained factor ${PR}_w$ exhibits a strong dependence on explanatory factors such as $P_{co}$ and ${PR}_c$ (76.2% and 77.6%, respectively). In turn, factors ${\mathrm{Y}}_{\mathrm{w}}$ and $P_{bf}$per have a moderate impact (32.5% and 43.0%, respectively). A very low correlation of factor $P_{w}per$ (2.9%) indicates inelastic demand for wheat; in other words, the relative change in demand for wheat changes to a lesser extent than the relative change in the price of wheat. The inelasticity of demand for wheat has also been demonstrated in studies by other authors [49,50]. The wheat production index per capita is an important measure because it takes into account the level of population growth and the level of feeding of the population on the planet; this is why we have argued for its inclusion in the constructed Model 1.

In turn, in Model 2, the explained factor ${PR}_c$ has a robust correlation with the explanatory factors $P_{co}$ and ${PR}_w$ (r = 0.74 and 0.77, respectively). On the other hand, factors ${\mathrm{Y}}_{\mathrm{c}}$ and $P_{bf}$per have a moderate impact on the explained factor (25.7% and 49.0%, respectively).

To test the hypothesis, we formalised the relationship between global wheat and corn prices (dependent variables) and selected explanatory factors by estimating two separate correlation–regression models—one for wheat and one for corn (

Table 5. Correlation coefficient matrix.

Model 1
Variables	${\mathit{P}\mathit{R}}_{\mathit{w}}$	${\mathit{P}}_{\mathit{c}\mathit{o}}$	${\mathit{P}\mathit{R}}_{\mathit{c}}$	${\mathit{Y}}_{\mathit{w}}$	${\mathit{P}}_{\mathit{w}}\mathit{p}\mathit{e}\mathit{r}$	${\mathit{P}}_{\mathit{b}\mathit{f}}$per
${PR}_w$	1.0000	0.8732	0.8806	0.3911	0.1695	0.4866
$P_{co}$	0.8732	1.0000	0.8645	0.3910	0.3001	0.5263
${PR}_c$	0.8806	0.8645	1.0000	0.6029	0.4054	0.7002
$Y_w$	0.3911	0.3910	0.6029	1.0000	0.7127	0.9447
$P_{w}per$	0.1695	0.3001	0.4054	0.7127	1.0000	0.5700
$P_{bf}$per	0.4866	0.5263	0.7002	0.9447	0.5700	1.0000
Model 2
Variables	${\mathit{P}\mathit{R}}_{\mathit{c}}$	${\mathit{P}}_{\mathit{c}\mathit{o}}$	${\mathit{P}\mathit{R}}_{\mathit{w}}$	${\mathit{Y}}_{\mathit{c}}$		${\mathit{P}}_{\mathit{b}\mathit{f}}$per
${PR}_c$	1.0000	0.8645	0.8806	0.5064		0.7002
$P_{co}$	0.8645	1.0000	0.8732	0.3592		0.5263
${PR}_w$	0.8806	0.8732	1.0000	0.3431		0.4866
$Y_c$	0.5064	0.3592	0.3431	1.0000		0.9352
$P_{bf}$per	0.7002	0.5263	0.4866	0.9352		1.0000

Source: Authors’ study.

).

The models developed for this study revealed the presence of collinearity, which, however, does not indicate economic dependencies. In particular, in both models, i.e., 1 and 2, factors ${PR}_w, {PR}_c$, and $P_{co}$ are not similar to each other in terms of value because their relationships result only from the general market situation, including the course of globalisation processes.

Both models indicate that the dependent and independent variables reinforce one another, revealing a higher degree of interdependence than suggested by the simple bivariate correlations reported above (Appendix A, Figure A1 and Figure A2). Model 1 confirmed a very high interdependence between the price of wheat and the price of crude oil (87.3%) and the price of wheat and the price of corn (88.1%). Factors ${\mathrm{Y}}_{\mathrm{c}}$ and $P_{bf}$per have a moderate relationship (39.1% and 48.7%, respectively), and the relationship of the factor $P_{w}per$ is 17.0%. Model 2 also confirmed a high correlation between the price of corn and the price of crude oil (86.5%), a high correlation between the price of corn and biodiesel production per 1 million people (70.0%), and a moderate correlation between the price of corn and its yield (50.6%).

4. Results

The regression models based on the selected indicators exhibit strong explanatory power, with coefficients of determination of R² = 0.89 for wheat and R² = 0.92 for corn (

Table 6. Empirical results: t-statistics (p-values).

Model 1		Model 2
Variable	${\mathit{P}\mathit{R}}_{\mathit{w}}$	Variable	${\mathit{P}\mathit{R}}_{\mathit{c}}$
Constant	1.608 (0.125)	Constant	2.453 * (0.024)
${\mathit{P}}_{\mathit{c}\mathit{o}}$	3.308 ** (0.004)	${\mathit{P}}_{\mathit{c}\mathit{o}}$	1.106 (0.282)
${\mathit{P}\mathit{R}}_{\mathit{c}}$	3.411 ** (0.003)	${\mathit{P}\mathit{R}}_{\mathit{w}}$	3.752 ** (0.001)
${\mathit{Y}}_{\mathit{w}}$	2.562 * (0.020)	${\mathit{Y}}_{\mathit{c}}$	−2.828 * (0.011)
${\mathit{P}}_{\mathit{w}}\mathit{p}\mathit{e}\mathit{r}$	−3.097 ** (0.006)	-	-
${\mathit{P}}_{\mathit{b}\mathit{f}}$per	−2.701 * (0.015)	${\mathit{P}}_{\mathit{b}\mathit{f}}$per	4.099 ** (0.001)
F-test (p-value)	32.077 ** (2.32 × 10−8)	F-test (p-value)	52.502 ** (3.22 × 10−10)
${\mathit{R}}^2$	0.899	${\mathit{R}}^2$	0.921
Number of observations	24	Number of observations	24
${\mathit{P}\mathit{R}}_{\mathit{w}}$$= 226.406 + 1.326 \times P_{co}$$+ 0.542 \times {PR}_c$$+ 0.171 {\times Y}_w$ + $(-6.267) {\times P}_{w}per$$+ (-0.669) {\times P}_{bf}$per		${\mathit{P}\mathit{R}}_{\mathit{c}}$$= 301.510 + 0.432{ \times P}_{co}$$+ 0.574 \times {PR}_w$ + $(-0.078) {\times Y}_c$$+ 0.818{ \times P}_{bf}\mathrm{p}\mathrm{e}\mathrm{r}$

Notes: * probability of error at 5% level; ** probability of error at 1% level. Source: Authors’ study.

).

Model 1 shows that the level of world wheat prices is strongly related to all independent variables in the specification, as evidenced by a large F-statistic (p < 0.05). All indicators exert statistically significant, proportional effects on wheat prices. Holding other variables constant, a 1 USD/barrel increase in the crude-oil price is estimated to raise the wheat price by 1.33 USD/t; a 1 USD/t increase in the corn price raises the wheat price by 0.54 USD/t. Conversely, a reduction in biodiesel production of 1 barrel per 1 million people is associated with a 0.67 USD/t increase in the wheat price. The coefficient $Y_w$ = 0.171 means that an increase in yield by 1 kg/ha causes an increase in the price of wheat by 0.17 USD/t, while an increase in the volume of wheat production by 1 kg per capita will cause a decrease in the price of wheat by 6.27 USD/t.

Model 2 shows a strong dependence of corn stock prices on wheat prices (${\mathrm{P}\mathrm{R}}_{\mathrm{w}}$ coefficient) and biodiesel production per capita ($P_{bf}$per coefficient), as well as a moderate negative dependence on corn yields ($Y_c$ coefficient), which is also confirmed by the high level of the F test (p-value). The $Y_c$ coefficient = (−0.078) means that as a result of a decrease in corn yields by 1 kg/ha, its price may increase by 0.078 USD/t. An increase in the price of crude oil by 1 USD/barrel will cause an increase in the price of corn by 0.43 USD/t. An increase in the price of wheat by 1 USD/t will cause an increase in the price of corn by 0.57 USD/t. An increase in world biodiesel production by 1 barrel per 1 million people causes a 0.82 USD/t increase in the price of corn.

In Model 1 for wheat, the regression shows a positive coefficient on the variable $Y_w$ (wheat yield): an increase in wheat yield leads to an increase in its price. Similar results were also obtained in earlier studies [19]. Conversely, corn yield negatively correlates with price $Y_c$ (β = –0.078), as increased output without proportional demand (especially for biofuels) leads to oversupply and falling prices [50].

The regression diagnostics confirm that the models are statistically reliable and appropriately specified, consistent with observed regularities at both micro- and macro-scales.

Globalisation strengthens price transmission across North American, European, Latin American, and Asian markets, helping to explain the close correlation between world wheat and corn prices. Because wheat and corn are, to a large extent, substitutes, an increase in the price of one tends to raise the price of the other. Rising oil prices elevate the costs of agricultural production, including cereals, underscoring the integration of energy and agricultural markets. In parallel, growing demand for biofuels—driven by decarbonisation policies—creates additional structural demand for corn.

Biofuels intensify competition for agricultural raw materials, especially corn. As the principal feedstock for bioethanol, corn attracts acreage when biofuel demand rises, prompting farmers to reorient cropping patterns (including shifts from wheat to corn) to maximise returns. Since wheat is not widely used for energy-oriented biomass, its role in meeting global energy demand is comparatively limited, implying a relative decline in its demand share within energy-driven grain markets.

The regression results exhibit high explanatory power (R² > 0.89), indicating a strong relationship between the selected explanatory variables and grain price dynamics. Wheat prices are significantly influenced by oil prices, corn prices, wheat yields, wheat production per capita, and biodiesel production, highlighting the multidimensional drivers of this market. Corn prices are shaped primarily by wheat prices, biodiesel production, and corn yields, reflecting both substitution effects between cereals and the influence of biofuel demand. The positive association between oil and grain prices is consistent with cost pass-through from the energy sector to agriculture. The negative effect of crop yields on prices accords with the classical supply–demand framework, whereby higher productivity lowers market prices. Biodiesel production plays a dual role—adding demand for corn while diminishing the relative importance of wheat in energy-oriented markets. Finally, the positive cross-price relationship between wheat and corn confirms their substitutability: increases in the price of one cereal tend to lift the price of the other.

5. Discussion

Globalisation intensifies price transmission across international markets, reinforcing interdependence between regions. The structural reorientation of agricultural production towards corn, stimulated by biofuel demand, reflects the growing integration of energy and agricultural markets.

In countries with subsidised bioethanol production, government policies—such as the Renewable Fuel Standard (RFS)—create additional demand for corn, redirecting agricultural investment away from wheat. Analyses for 2021 indicate that the RFS increased annual ethanol output in the United States by about 5.5 billion gallons, requiring an additional 1.3 billion bushels of corn and raising corn prices by 31% and wheat prices by approximately 20%, thereby incentivising shifts in acreage [51].

Overall, the empirical findings confirm that grain price formation is strongly conditioned by both agricultural fundamentals and global energy-market dynamics—factors that should be incorporated into future forecasting and policy design.

Greater corn-based biofuel production generates more ethanol by-products—distillers’ dried grains with solubles (DDGS)—which are increasingly used in the feed industry in place of wheat [52]. Consequently, demand for feed wheat declines. Global processors have shifted towards more affordable and readily available raw materials—primarily corn—for feed, alcohol, and biofuels [53]. Wheat remains a relatively more expensive food crop, limiting its role in the bioeconomy segment and narrowing its demand channels to the food market.

The resulting asymmetric yield effect—higher yields are positively correlated with wheat prices but negatively with corn—can be explained by a combination of agronomic signals and market-economic feedbacks. First, rising wheat yields may act as an indicator of sectoral efficiency and structural improvement rather than merely increasing short-run physical supply. Higher wheat prices raise expected returns and thus stimulate investments in improved varieties, fertilisation, mechanisation and precision agronomy; these investments raise yields over time, producing a positive long-run association between price and yield (price → investment → higher yields). Empirical studies show that price signals often precede adoption and yield responses, implying causality from price to productivity improvements in cereals (e.g., Xie & Wang find price changes Granger-cause yield changes in agricultural contexts) [54].

Second, the wheat market’s institutional and demand structure can moderate the classical supply effect. Wheat is used across food, feed, and industrial channels with flexible substitution possibilities (including feed wheat ↔ corn), and storage/marketing behaviour (strategic stocks) can produce situations where price increases anticipate yield responses rather than simply reflect contemporaneous oversupply [45,55]. Long-run cointegration studies support the notion that price dynamics and yield adjustments are interlinked with lagged adjustment processes (so price–lead relationships are plausible) [54].

Third, the difference can reflect sectoral heterogeneity in adoption and responsiveness to modern technologies. Corn production, particularly in major producing countries, has already absorbed extensive precision-agriculture innovations that boost yields at scale [56]; when yields expand rapidly (e.g., favourable weather, strong technology uptake), the supply response can be large and swift, exerting downward pressure on prices if demand (including biofuel off-take) does not follow [1]. Conversely, wheat production systems in many regions still have scope for productivity gains from technology adoption [57,58], so price increases more often translate into investment and gradual yield growth rather than immediate oversupply. Recent studies show accelerating uptake of digital and precision technologies that raise productivity but also change short-run supply dynamics differently across crops and regions [59].

Climate change indirectly affects grain price fluctuations through yield outcomes. Both wheat and corn exhibit substantial yield gaps, especially in developing countries [57,60]. The cost-effectiveness of closing these gaps depends on local water and nutrient management, the production system, regional conditions, and the technologies adopted. In some regions, increased fertilisation and irrigation have delivered high benefits at moderate costs, whereas elsewhere the same techniques have not proved cost-effective [61].

Investing in new technologies and practices can raise yields and improve resource-use efficiency, but it typically entails substantial up-front capital outlays and adoption costs. The payback period and risk profile are context-dependent, which can constrain uptake without targeted finance, risk-management instruments, or policy incentives.

The integration of energy and agricultural markets—demonstrated through rigorous econometric modelling—offers critical insights for policymakers who must address food-price inflation as a function of energy and biofuel interventions, underscoring the importance of coordinated energy-agriculture strategies. Studies on energy policy impacts on food prices highlight the fact that even small energy price adjustments can significantly affect food costs and consumer welfare, particularly in low-income regions [62]. Understanding price transmission between global energy and grain markets also supports macroeconomic stability efforts; the U.S. Bureau of Labor Statistics notes that surging grain prices can ripple through the broader economy, amplifying food inflation and reducing affordability [63]. For political economists, articulating how biofuel mandates—like the U.S. Renewable Fuel Standard—artificially elevate corn prices (and indirectly wheat) helps explain the electoral and lobbying dynamics governing such policy shifts [64].

In environmental and energy economics, knowledge of grain—energy interlinkages informs the design of incentive instruments that avoid unintended externalities, such as food shortages or land-use shifts, aligning with the ‘induced innovation’ framework that emphasises policy-driven technological and market responses [65]. International development agencies and food security planners benefit from these findings, as food systems are deeply affected by energy shocks, which—if unmodeled—can exacerbate crises in vulnerable regions [62,63].

Financial market stakeholders—such as commodity traders, risk managers, and agribusiness firms—also gain from better forecasting of grain prices that incorporate biofuel and oil market dynamics, enabling more accurate hedging strategies. Futures markets for grains, widely used for risk mitigation, depend on understanding such cross-market influences [66].

Finally, from an academic standpoint, these results contribute to the literature on globalization and price transmission, offering empirical evidence that supports theories of market integration and substitutability in international agricultural systems [67].

6. Conclusions

Agricultural markets exhibit pronounced price volatility that is increasingly shaped by energy policy and industrial strategies rather than purely agronomic conditions. Analysis of wheat and corn prices over 2000–2023 confirms this pattern, with energy-market variables and biofuel-related measures exerting significant and persistent effects alongside yields and other fundamentals.

Econometric models demonstrate strong explanatory power (R² = 0.899 for wheat; R² = 0.921 for corn), with prices influenced by oil prices, substitute cereals, and biofuel production variables. Importantly, yield impacts are asymmetric: wheat yields positively correlate with price (β = 0.171), reflecting sector efficiency, while corn yields show a negative relationship (β = –0.078), consistent with oversupply effects. Wheat and corn prices correlate strongly owing to substitutability in global markets. Globalisation and market integration amplify price transmission, while rising oil prices add cost pressures that raise grain prices. Although excluded from the regressions due to collinearity, climatic factors were found to be strongly correlated with yields and biofuel production, underscoring the importance of climate-related risks.

The econometric analysis demonstrates that grain price dynamics cannot be explained solely by agronomic variables; they are strongly conditioned by energy-market fluctuations and biofuel policies. By integrating these cross-sectoral linkages, the models provide a more comprehensive framework for understanding price formation in global agricultural markets.

These results highlight the need to design agricultural and energy policies that explicitly account for cross-market interactions. Rising oil prices impose cost pressures that propagate through food systems, suggesting the importance of integrated energy–agriculture modelling frameworks. Biofuel policies, such as the US Renewable Fuel Standard, structurally increase corn demand and reallocate land from wheat, while by-products like DDGS alter feed-market balances. Such dynamics illustrate how policy choices in one sector reverberate through others, reshaping price formation and competitiveness. Adaptation and resilience-oriented agricultural policies are also essential to buffer the variability caused by droughts, temperature increases, and rainfall shifts. More advanced dynamic system models, including computable general equilibrium (CGE) frameworks, can be valuable tools for policymakers to simulate the broader effects of energy shocks and climate variability on food prices.

Overall, this study advances understanding of global grain price dynamics by demonstrating the asymmetric yield effects and the dominant role of energy-related drivers. This dual perspective strengthens the academic debate on market integration and provides a practical foundation for developing policies that enhance food security and macroeconomic stability in the context of global energy transitions. Future research should incorporate climate and geopolitical variables to improve explanations of long-run volatility and stability in global grain markets.