Abstract This study contributes to the industrial organization literature on vertical coordination and contract governance in agri-food systems by providing a multi-dimensional empirical assessment of a processing tomato supply chain in Upper Northeastern Thailand, where smallholder farmers engage processors through contract farming arrangements. Building on transaction cost economics and principal-agent theory, a comprehensive measurement framework comprising 56 performance indicators organized across nine dimensions was developed and applied to 346 contracted farmers across four provinces during the 2024–2025 production season. Stochastic frontier analysis (SFA) was employed to estimate farm-level technical efficiency, while Spearman rank correlation analysis examined linkages between relational governance dimensions and farm-level outcomes. Results reveal substantial vertical coordination failures, particularly in joint decision-making (mean = 2.08 ± 1.09) and problem-solving (mean = 1.77 ± 1.11), alongside pronounced market power asymmetries where farmers report high processor leverage (RBA9 = 3.49) but limited bargaining capacity (RBA8 = 2.46). Traceability infrastructure is underdeveloped (only 31.13 % of farmers). Mean technical efficiency is 69.10 %, indicating a 30.9 % output gap attributable to managerial and agronomic constraints, with substantial provincial variation ranging from 55.47 % in Nakhon Phanom to 89.86 % in Buengkan. The Cobb-Douglas production frontier reveals increasing returns to scale (Σβ = 2.015), while farming experience, furrow irrigation, and harvest frequency significantly reduce technical inefficiency. Critically, information sharing (r s = 0.124, p = 0.021) and resource allocation fairness (r s = 0.114, p = 0.034) are significantly associated with farm profit but not with technical efficiency, establishing a dual-pathway finding: relational governance improves economic outcomes through market-side mechanisms such as improved pricing, reduced transaction costs, and more equitable rent distribution whereas technical efficiency is driven by agronomic factors orthogonal to governance quality. Policy recommendations address contract design reform, producer organization development, targeted extension services, traceability infrastructure investment, and competitive market structure oversight.
Abstract Despite heavily relying on and employing more than half of its total labour force in the agricultural sector, Sub-Saharan Africa (SSA) experiences the highest levels of food insecurity and the highest undernourished children in the world, with rising levels of hunger. The purpose of this paper was to investigate the agricultural sector effects of environmental and institutional quality in SSA from 2002 to 2021. Data was analyzed using the Fixed Effects estimator with Driscoll Kraay Standard Error approach. The key findings reveal that ceteris paribus a 1 % increase in CO2 emissions associated with 0.366 %, 0.244 % and 0.198 % reduction in agriculture value added (AVA) in the low-income (LIC), lower-middle-income (LMI) and upper-middle-income (UMI) categories respectively. The corresponding reactions from a 1 % increase in energy intensity are 0.556 %, 0.174 % and 0.37 % for the respective income groups. A disaggregation of the agricultural production showed that the crop, food and livestock production models suffer differentially from increases in these environmental quality indicators. On the other hand, a 1-unit increase in institutional quality corresponds to increases of about 5.1 %, 7.8 % and 13.8 % in AVA in LIC, LMI and UMI respectively. In terms of size, CO2 emissions impact largest the LIC, whereas energy intensity and institutional quality affect largely the UMI group, considering all the agricultural indicators. While higher energy intensity worsens the adverse impact of carbon emissions on agriculture, stronger institutional quality weakens it. Policies to curb carbon emissions, reduce energy intensity, and strengthen institutional quality would benefit AVA, FPI, CPI and LPI.
Abstract Research on the determinants of coffee prices is scarce. In this study, we examine how coffee prices are associated with two key coffee attributes – taste quality and roast level – within a price hedonic framework. We report three main findings. First, taste quality has a positive but non-linear effect on price: Among coffee types below the threshold for superior quality, an additional quality point has only a modest effect, whereas among superior-quality coffees, the same increase in quality results in a substantially larger price premium. Second, roast level has a positive and roughly linear effect on price, indicating that lighter roasts are more expensive than darker ones. Third, both the non-linear quality effect and the linear roast-level effect vary across the coffee price distribution, with substantially larger effects among the most expensive coffees than among the least expensive. Although these findings show how quality and roast level shape price formation in the specialty coffee market, their novel character also suggests that more research is needed on the determinants of coffee prices.
Abstract Food industries drive economic growth, yet their substantial within-sector variation remains poorly understood. This study examines heterogeneous socio-economic impacts across nine food industry sectors using continuous difference-in-differences methodology applied to panel data from Catalonia (2015–2022), covering 64 industries economy-wide. Our continuous treatment captures establishment growth intensity, enabling sector-specific impact estimation. The analysis reveals high heterogeneity: Other Food Products generates employment increases substantially above the economy-wide average, while Dairy Products exhibits large negative effects. Spillover analysis was inconclusive due to severe multicollinearity among related sector measures, leaving cross-sector transmission as an open question. Value chain analysis reveals upstream suppliers generate significant multiplier effects across downstream industries. Robustness checks – including alternative weighting schemes, COVID-period sensitivity analysis, and export intensity controls – confirm that the observed heterogeneity is structural rather than driven by measurement choices or external shocks. These findings suggest that uniform industrial policies may be inadequate for the food industry, highlighting the potential value of differentiated sector-specific approaches. The methodological framework provides a replicable template for examining heterogeneous Industry 4.0 impacts across industries and regions.
Abstract Understanding how price signals are transmitted along agricultural value chains is critical for assessing market coordination and equity, particularly in developing economies. This study analyzes vertical price transmission in the Philippine corn market across farmgate, wholesale, and retail levels using monthly data from 1990 to 2020. Employing both linear and nonlinear econometric approaches – specifically the Threshold Vector Error Correction Model (TVECM) – the analysis identifies cointegrated relationships among prices and reveals asymmetric, threshold-dependent adjustments. Results show that farmgate prices respond strongly to deviations from equilibrium, wholesale prices adjust in a regime-dependent manner, and retail prices remain relatively rigid. These dynamics indicate that price transmission is often sluggish and uneven, reflecting transaction costs, market power disparities, and coordination frictions. Such patterns place a disproportionate adjustment burden on farmers while limiting consumer benefits from falling upstream prices. The findings highlight the importance of policies that improve supply chain integration, strengthen farmer bargaining power, and enhance price monitoring. By focusing on asymmetric transmission and nonlinear adjustments, this study provides new insights into how price relationships shape the performance of the Philippine corn market.
Abstract Agri-food systems are responsible for a substantial share of global carbon dioxide (CO 2 ) emissions, with food waste constituting one of the most avoidable yet underexplored sources of emissions. Existing studies largely approach waste-related emissions from technical or accounting perspectives, offering limited insight into their strategic foundations. This study aims to examine waste-related CO 2 emissions in agri-food systems from a strategic management perspective, identifying distinct emission configurations across countries. The study adopts a configuration-based research design, using cross-country agri-food CO 2 emissions data from 236 countries for the period 1990–2020. Cluster analysis is applied to country-level averages of waste-related CO 2 emissions, total agri-food system emissions, and the relative share of waste emissions. The analysis is theoretically grounded in the Natural Resource-Based View (NRBV), which conceptualizes environmental performance as a strategic outcome. The results reveal three distinct waste-related emission configurations that differ significantly in both emission scale and composition. These configurations can be interpreted as outcome patterns consistent with different sustainability orientations, rather than as purely technological or structural differences. This finding highlights the potential strategic relevance of waste-related emission structures within agri-food systems. By linking emission configurations with the NRBV framework, the study contributes to the sustainability and strategic management literature by offering a configuration-based perspective on waste-related CO 2 emissions in agri-food systems. These findings provide insights for both sustainability research and policy discussions aimed at reducing waste-related emissions in agri-food systems.
Abstract The consolidation of US dairy farming has led to concentrated manure production, resulting in manure management challenges and GHG emissions from unmanaged manure. Policies such as subsidizing compost production and implementing strict Nutrient Application Standards incentivize composting of manure and mitigate GHG emissions while impacting compost demand and herd size. Our model indicates that both policies would increase compost manure production and decrease compost price which would reduce unsold fresh manure, and cut emissions. However, when considering the size of the farm, subsidizing compost production and strict Nutrient Application Standards have different impacts. Nutrient Application Standards increase the cost of managing excess manure, leading to smaller herd size whereas, the subsidy has no impact on herd size as long as the farm retains unsold manure.
Market power represents distortions that hinder the efficient allocation of resources within an economy. The significance of measuring such distortions is particularly pronounced in Philippine agriculture which has been subject to periodic food affordability crises. This paper is the first to directly quantify markups and assess returns to scale for agricultural suppliers and manufacturers. Using establishment-level surveys for Philippine agricultural suppliers and manufacturers from 2012 to 2018, I estimate a national average markup parameter for suppliers and manufacturers through production function regressions. I use panel data models to consistently estimate output elasticities for materials’ expenditure with which I back out theory-consistent markups and scale economies separately for agricultural suppliers and manufacturers. For suppliers, I find average markups of 8 % and constant returns to scale. For agricultural manufacturers, I find markups above 50 % and modest but statistically significant increasing returns to scale of 11 %. These estimates are consistent with the theoretical relationship between markups, profits, and returns to scale. Estimates of fixed costs over time inform us that markups are a more important factor keeping profit rates elevated for manufacturing than for upstream suppliers.
Public interest in the pork sector renewed in the wake of COVID-19 disruptions. Policy proposals addressed a range of issues – real and perceived – in meatpacking and the pork sector. Evaluating these policies credibly requires accurately representing the industry structure. One feature of the pork supply chain is the use of alternative marketing arrangements to procure hogs. This study seeks to augment existing literature by fully accounting for heterogeneity across alternative marketing arrangements. A structural econometric model links hog supply and pork demand through a representative packer’s optimal procurement decision of hogs on the spot (negotiated) market. Prices are then discovered for alternative marketing arrangements according to price rules. A market power conjecture allows for testing the source and extent of the representative packer’s market power. Results for the period from 2013 to 2024 indicate a lower degree of market power than found in the existing literature and differential contributions from each type of alternative marketing arrangement. The share of hogs procured via methods from which packers derive market power has consistently grown in for the past two decades, however, so the degree of market power may increase if procurement trends continue.
The main goal of this study was to create a strong predictive model for forecasting olive oil prices. To do this, we applied four machine learning models in Python: Random Forest, Gradient Boosting, Decision Tree, and Support Vector Regression, using 164 monthly price observations along with factors like temperature, precipitation, consumer price index, IBEX35 stock market prices, EUR/USD exchange rate, and import/export quantities. The results showed that Random Forest and Gradient Boosting models performed the best. The Spearman correlation analysis revealed that the exchange rate had a strong negative correlation with prices, while the consumer price index and import quantity had moderate positive correlations. Random Forest highlighted the consumer price index as the most important factor in predicting olive oil prices. This study fills a gap in existing research and provides practical insights for companies in the olive oil industry to better monitor and forecast prices, helping with profitability, risk management, stock optimization, and investment decisions.
The objective of this study is to examine the dynamic adjustment of retail prices following changes in input prices within an oligopolistic and vertically non-integrated market. Data are weekly retail and wholesale prices between February 2010 and August 2013. The methodology employed is the Nonlinear Autoregressive Distributed Lag (NARDL) model, which captures both short- and long-run price dynamics. The empirical findings reveal evidence of the “rockets and feathers” hypothesis for more than half of the products examined, indicating an inflationary impact on retail prices, thus a temporal or permanent decrease in consumer welfare. This result is in contrast with the perishable nature of fruits and vegetables.
I explore the relationship between incentive alignment and effort provision in the canonical multi-task principal-agent contract model. I rely on an experiment where human subjects act as agents making effort decisions based on contract structures providing both theoretically-correct and distorted incentives. Results suggest that distorted contracts, offering weaker incentives than theory recommends, may be less detrimental than expected when performance and quality are well-aligned. The experiment also confirms the theoretical prediction that strong incentives, when misalignment is severe, can increase effort but reduce the principal’s profit. The findings contribute to understanding the prevalence of contracts with weaker-than-optimal incentives, suggesting that such incentive schemes may be more prevalent in the real world when the cost of providing less-than-ideal incentives is relatively low.
We provide empirical evidence to explain the apparent discrepancy between consumption choices and voting outcomes for a recent high-profile animal welfare case: California’s ban of eggs produced with caged hens. The model juxtaposes the private good aspect of buying decisions with the public good aspect of voting, and yields testable propositions for the vote-buy gap. These implications are evaluated in a revealed-preferences setting using a novel combination of voting and egg-purchase data. Results show that the vote-buy gap depends on the egg price differential, and the distribution of consumers’ heterogeneous preference for animal welfare issues.
A common form of cooperation in rural areas corresponds to informal alliances in which farmers save productive costs, for example, by sharing inputs, machinery, and information. By reducing costs cooperatively, these alliances contribute to agricultural land management by increasing efficiency. Despite the benefits, these alliances are typically formed by a small number of farmers. Some researchers attribute this lack of participation to distrust. To model this type of cooperation, this article extends the traditional Network of Collaboration approach which has traditionally been developed in an oligopolistic context. While oligopoly is a type of market structure that can exists in agriculture, most of the farmers who participate in informal cooperation are small enterprises suggesting that the price-taking assumption is a more realistic description for these farmers. We use this market structure extension to show that profit-maximising farmers consider the positive impact of cost reduction; the negative impact of distrust; and the expected gain in output when joining new informal alliances. As such, recommendations are provided to integrate these three factors to facilitate the formation of beneficial informal network alliances in the rural landscape and their positive externality in agricultural land management.
Structural economic transformation theories have emphasized the role of agricultural development in industrialization and vice versa. The question is whether industrial development and agricultural development were complementary. Using unique data from Sweden (1800–2020), we investigate the causal link between agriculture and industry. In the presence of structural shifts and breaks, the causal direction is established using the Fourier Granger causality technique. Based on the results, Sweden supports the growth complementarity hypothesis. This indicates that sectoral convergence tendencies are encouraged by the larger expansion of industry, which frequently spreads to agriculture and vice versa.
The main goal of this study was to create a strong predictive model for forecasting olive oil prices. To do this, we applied four machine learning models in Python: Random Forest, Gradient Boosting, Decision Tree, and Support Vector Regression, using 164 monthly price observations along with factors like temperature, precipitation, consumer price index, IBEX35 stock market prices, EUR/USD exchange rate, and import/export quantities. The results showed that Random Forest and Gradient Boosting models performed the best. The Spearman correlation analysis revealed that the exchange rate had a strong negative correlation with prices, while the consumer price index and import quantity had moderate positive correlations. Random Forest highlighted the consumer price index as the most important factor in predicting olive oil prices. This study fills a gap in existing research and provides practical insights for companies in the olive oil industry to better monitor and forecast prices, helping with profitability, risk management, stock optimization, and investment decisions.
Abstract The present study investigates the linkages among the futures prices of feeder cattle, live cattle and lean hogs in the US. This has been pursued using a flexible methodology that allows modelling price relationships at different parts of their joint distribution. Data are daily closing prices for the period between 1/1/2015 and 12/31/2023. According to the empirical results: i) livestock commodities boom together and crash (with one exception) together, ii) extreme price decreases are transmitted with higher intensity compared to extreme price increases, iii) transmission asymmetries in prices, between livestock commodities, can occur at the tails as well as at the median of the joint distributions. Lastly, opportunities for speculators to profit from the spread between the commodities of feeder cattle and live cattle can be present.
Abstract This paper discusses international trade prospects of the main wine-producing countries over different time horizons. Our framework is composed of two approaches and aims at renewing and extending the findings of the existing literature on the dynamics of major wine exporters. Firstly, we apply a Porter’s diamond analysis in order to disentangle competitive advantages in 16 countries from various features of the global wine market. Then, we assess short- and medium-term prospects from the evolution of past trends. Secondly, we use data from a survey of (57) wine experts as a robustness analysis to complete the prospective dimension over the long term. Compared with previous literature (prior to the Covid and Ukraine shocks), our results show that the international competitiveness of countries has significantly changed over the last five years. France and Italy should maintain their leadership, but Spain and Chile are slipping back, while China could collapse. New Zealand could eventually become a major leader. These results indicate that competitive advantages are very dynamic in the wine industry, despite static natural endowments (excluding climate). Investment incentives and innovation should therefore play a key role in the long term in this sector, which is often presented as being driven by tradition and natural factors.
Abstract In this paper we investigated the dynamic of competition in the U.S. broiler chicken industry from 2000 to 2023, focusing on the detection and the implications of alleged collusive behaviour. By employing a structural econometrics approach, we estimated the conjectural elasticities, offering an explanation for how potential collusive arrangements evolved over time. Our most important results show that the alleged collusion period elevated the conjectural elasticities for all defendants but increased the conjectural elasticity of the group of three largest publicly traded broiler companies the most. These results suggest that the industry tacit collusion could have morphed into explicit collusion, with the largest three firms being the likely leaders of the conspiracy. In addition, the computed Lerner indexes show that the market power has increased during the collusion period relative to the pre-collusion period and remained elevated afterwards, suggesting stubborn impact of collusion on competition even during the post collusion period.
Abstract The consumption of invasive species could be an opportunity to regulate these species to limit the negative environmental impacts. However, the commercialisation of an exogenous species raises several questions. We assess the acceptance of wels catfish, an invasive species in an alpine lake, and the willingness to pay (WTP) for environmentally friendly and locally produced wels catfish products. The results show that knowledge is an important dimension to explain consumption. Informing consumers about an environmental or local dimension increases WTP for wels catfish, but it does not increase product acceptance. There is no additional premium when both types of information are used together.
