Search

Price instability is recognized as a major driver of food insecurity that may have negative impacts on the social, economic and health status of many households. Since the food price spikes of 2007–8, prices of the main food and feed crops have shown high variability in the international and local markets for almost a decade. Although agricultural commodity prices are influenced by many factors, shocks on agricultural productions are often recognized as a major driver of price volatility. Adverse environmental conditions and the resulting decline in regional production have be shown to have greatly contributed to spikes in global food prices. The connection between production levels, agricultural commodity prices and food security has led many companies, governments and international organizations to develop forecasting systems to anticipate production shortages or, on the contrary, overproduction. The last decade has seen a rising number of applications of machine learning, with a particular focus on anticipating agricultural production shocks. The reasons are multiple. First, machine learning techniques can now be more easily applied than ever before using specialized packages implemented in open source packages. Second, large agricultural databases are becoming increasingly easily accessible, thanks in large part to the emergence of good practices in open science. Finally, mechanistic models were found to miss several major yield losses in Europe and in the US which led scientists to look for alternative solutions. For all these reasons, different machine learning tools are now being used more often to analyse production shocks, identify their origins, and assess their consequences on food security. Based on a brief review of the literature, we identify the main biotic and abiotic factors determining crop production shocks, propose a hierarchy for the factors with the most impact, and show that these factors are often interdependent, making crop yield forecasting rather difficult. Next, based on our own recent research projects, we show how the use of machine learning and of probabilistic models could open – in connection with the development of open access databases and new powerful algorithms – a new avenue for predicting production shocks and their impacts on agricultural prices. Lastly, we discuss the economic mechanisms that link together seasonal production forecasts and market impacts and how improving production forecasts could reduce price volatility and the risk of abrupt shifts in price trends., Sur la base d'une bref revue de littérature, nous identifions les principaux facteurs biotiques et abiotiques qui déterminent les chocs de production en agriculture, nous proposons une hiérarchie des factors ayant le plus d'impacts, et nous montrons que ces facteurs sont souvent interdépendant, ce qui rend la prédiction des rendements assez difficile. Ensuite, nous montrons comment l'usage d'outils d'apprentissage machine et de modèles probabilistes, en lien avec le développement de bases de données ouvertes et de nouveaux algorithmes, pourraient ouvrir de nouvelles perspectives pour la prédiction des chocs de production et leurs impacts sur les prix agricoles. Enfin, nous discutons les mécanismes économiques qui lient ensemble prévisions saisonnières de production et impacts de marché, et comment améliorer les prévisions de production peut réduire la volatilité des prix et le risque de mouvements de prix brutaux.

How scientists might better contribute to SDGs in the land sector? At the invitation of INRAE, a group of high-level international experts delivers a comprehensive and systemic approach.

This paper develops and estimates a rational expectations storage model with rich dynamic features. The model incorporates elastic supply, long-run demand and cost trends, and four structural shocks. It is estimated by indirect inference using a linear supply and demand model as an auxiliary model. This approach deviates from the common practice of estimating storage models only on prices and allows all parameters to be estimated. The estimation is carried out on a market representing the caloric aggregate of four basic staples: maize, rice, soybean, and wheat. The results show that our extended storage model is consistent with most of the moments in the data, including the high price autocorrelation which is the subject of longstanding debate. However, it fails to replicate the strength of the negative correlation between consumption and price.

This paper shows how gains from trade are conditioned by love of variety, defined as the extent to which an additional product variety generates benefits in either final or intermediate consumption. We develop a multi-country, multi-sector gravity trade model where love of variety is parameterized separately from product substitutability using a generalized CES demand function, and show analytically how gains from trade depend on love of variety through different channels that we identify and interpret. In this context, except for very specific parameterizations, gains from trade differ between a heterogeneous- and a homogeneous-firm model. Counterfactual simulations based on a calibrated version of this model show that, all other things being equal, the assessed gains from trade commonly vary by a proportion of one to three depending on the value of the love-of-variety elasticity, in a way that differs significantly across countries. Trade war simulations also point to the strong sensitivity of the assessed impacts. We conclude that love of variety is a key determinant of the gains from trade, an aspect that has so far been overlooked for the sake of convenience in the modeling framework and due to lack of empirical estimates.

First online; This article provides a framework to estimate the potential effects and benefits of the provision of market information in storable commodity markets. This framework is applied to the case of production forecasts for the soybean market. A rational expectations storage model of the global soybean market accounting for both inter-annual and intra-annual market dynamics is built. Shocks that occur between planting and harvesting affect the size of the potential harvest. Estimates of the size of these shocks are reported publicly, and affect the market equilibrium through adjustments to stock levels. By varying counterfactually the observability of seasonal shocks, we can estimate the efficiency gains related to the availability of advance information. They are equivalent to 2% of storage costs; the reduction of stock levels being the main channel explaining the welfare gains. The presence of advance information has a limited effect on inter-annual price volatility but redistributes price volatility during the season, increasing it just before harvest when almost all news has been received and stocks are tight, and decreasing it after harvest. The effect of news shocks is stronger on higher-order moments of the distribution with a strong decrease in skewness and kurtosis related to the lower frequency of price spikes.; Cet article propose un cadre d'analyse des effets potentiels et des bénéfices de la fourniture d'information de marché dans les marchés de matières premières stockables. Ce cadre est appliqué au cas des prévisions de production pour le marché du soja. Un modèle de stockage à anticipations rationnelles du marché mondial du soja prenant en compte à la fois les dynamique intra- et inter-annuelles est construit. Des chocs arrivant entre les semis et la récolte affectent la taille de la récolte potentielle. Des estimations de la taille de ces chocs sont rendues publiques et affectent l'équilibre de marché via l'ajustement des niveaux de stocks. En variant l'observabilité des chocs, nous pouvons estimer les gains d'efficiences en lien avec la disponibilité de l'information. Ils sont équivalents à 2% des coûts de stockage; la réduction des niveaux de stocks étant la principale explication des gains de bien-être. La présence d'information avant la récolte a un effet limité sur la volatilité des prix inter-annuelle mais redistribue la volatilité des prix durant la saison, l'augmentant juste avant la récolte lorsque presque toute les nouvelles ont été reçues et les stocks sont limités, et la diminuant après la récolte. L'effet des chocs de nouvelles est plus important sur les moments d'ordre élevé avec une forte décroissance de l’asymétrie et du kurtosis en lien avec la plus faible fréquence des pics de prix.

Presentation ID 16311

This study provides a framework to estimate the potential effects and benefits of the provision of market information in storable commodity markets. This framework is applied to the case of production forecasts for the soybean market. A rational expectations storage model of the global soybean market accounting for both inter-annual and intra-annual market dynamics is built. Shocks that occur between planting and harvesting affect the size of the potential harvest. Estimates of the size of these shocks are reported publicly, and affect the market equilibrium through adjustments to stock levels. By varying counterfactually the observability of seasonal shocks, we can estimate the efficiency gains related to the availability of advance information. They are equivalent to 2% of storage costs; the reduction of stock levels being the main channel explaining the welfare gains. The presence of advance information has a limited effect on inter-annual price volatility but redistributes price volatility during the season, increasing it just before harvest when almost all news has been received and stocks are tight, and decreasing it after harvest. The effect of news shocks is stronger on higher-order moments of the distribution with a strong decrease in skewness and kurtosis related to the lower frequency of price spikes.

This paper presents a new strategy to estimate the rational expectations storage model. It uses information on prices and quantities – consumption and production – in contrast to previous approaches which use only prices. This additional information allows us to estimate a model with elastic supply, and to identify parameters such as supply and demand elasticities, which are left unidentified when using prices alone. The estimation relies on the Bayesian methods popularized in the literature on the estimation of DSGE models. It is carried out on a market representing the caloric aggregate of the four basic staples – maize, rice, soybeans, and wheat – from 1961 to 2006. The results show that to be consistent with the observed volatility of consumption, production, and price, elasticities have to be in the lower ranges of the elasticities in the literature, a result consistent with recent instrumental variable estimations on the same sample.

This thesis for the accreditation to supervise research presents my work from 2011 to 2016 dealing with agricultural market dynamics. I have first considered issues of short-run dynamics through two questions: the study of food price stabilization policies and the structural econometrics of the storage model. Regarding stabilization policies, I have analyzed the justifications for public intervention, the different types of policies (storage, trade policy) and how they should be combined, and the question of international disciplines, because these policies tend to be non-cooperative. The storage model is central to my work on stabilization policies, but its empirical performances are still debated. So, I have tried to contribute to the recent renewal of its empirical analysis. To this end, I have developed two new estimation strategies. In a first article, I have demonstrated the importance of accounting for the possibly non-stationary behavior of commodity prices when estimating the storage model. In a second article, I have proposed an estimation method using the information available in quantities, in addition to those available in prices. Recently, I also became interested in issues of long-run dynamics, until now through the angle of the dynamics of food demand. I have proposed the first economic model of the nutritional transition at the global level. This microfounded model is estimated and allows to capture the main features of the nutritional transition: increase of calories with income per capita, decrease of starchy staple food consumption, diet diversification towards other food groups in particular sugar, fats, and animal-based products.; Ce mémoire d'Habilitation à Diriger des Recherches présente mes travaux de 2011 à 2016 portant sur la dynamique des marchés agricoles. Je me suis tout d'abord intéressé aux dynamiques de court terme des prix agricoles à travers deux questions: l'étude des politiques de stabilisation des prix alimentaires et l'économétrie structurelle du modèle de stockage. Concernant les politiques de stabilisations, j'ai analysé les justifications à une intervention publique, les différents types de politiques (stockage, politique commerciale) et comment elles doivent s'articuler, et le problème des disciplines internationales, car ces politiques tendent à être non coopératives. Le modèle de stockage est au cœur de mes travaux sur les politiques de stabilisation, mais ses performances empiriques étant encore débattues, j'ai donc essayé de contribuer à un renouvellement de son analyse empirique. Pour cela, j'ai développé deux nouvelles approches d'estimation. Dans un premier article, j'ai montré l'importance de prendre en compte la non-stationarité des prix des commodités dans l'estimation du modèle de stockage. Dans un deuxième article, j'ai proposé une méthode d'estimation du modèle de stockage utilisant les informations contenues dans les quantités, en plus de celles provenant des prix. J'ai récemment commencé à m'intéresser aussi aux dynamiques de long terme des marchés agricoles, jusqu'ici via le prisme de l'évolution de la demande alimentaire. J'ai proposé le premier modèle économique représentant la transition nutritionnelle au niveau international. Ce modèle microfondé est estimé économétriquement et permet de capturer les grandes caractéristiques de la transition nutritionnelle: augmentation des calories avec le revenu par tête, diminution de la part des féculents, diversification de la diète vers les autres groupes nutritionnels en particulier les sucres, les graisses et les produits animaux.

Le retour d’une importante volatilité des prix agricoles au cours de la dernière décennie a relancé le débat sur les politiques de stabilisation des prix. Cette synthèse fait le point sur ce débat en se basant essentiellement sur les travaux issus des recherches du département SAE2 de l’Inra. Cette synthèse tente de répondre à trois questions. (1) Faut-il intervenir avec des politiques de stabilisation des prix ? Cela implique, dans un premier temps, de s’intéresser à l’optimalité de la dynamique des prix et donc de revenir sur le débat ancien entre modèles à dynamique endogène (cobweb) et modèles à anticipations rationnelles. Dans un second temps, se pose la question de la capacité des agents à faire face au risque de prix et aux bénéfices qu’ils pourraient retirer d’une stabilisation des prix. (2) S’il est justifié d’intervenir, comment le faire ? Il s’agit là d’analyser la conception des politiques de stabilisation : la meilleure manière d’intervenir en combinant politique de stockage et politique commerciale pour un objectif donné. Ces interventions optimales sont ensuite mises en perspective par rapport aux politiques effectivement utilisées dans le passé. (3) Comment limiter les effets déstabilisateurs de ces politiques sur le plan international ? Des politiques qui font sens du point du vue d’un pays peuvent en effet être dommageables aux pays tiers. Pour limiter les effets négatifs, il est important de comprendre dans quelle mesure ces politiques peuvent faire l’objet d’accords internationaux. Pour l’instant, les conclusions de la recherche sont plutôt pessimistes : une coordination internationale sur des politiques commerciales contra-cycliques moins interventionnistes est peu probable.

Paper #9043