Economic efficiency measures and its determinants for irrigated wheat farms in Tunisia: a DEA approach

francaisL'objectif de cet article est double. Dans un premier temps, les scores d'efficacite technique, allocative et economique d'un echantillon de 170 exploitations cerealieres de la region de Chebika (Centre de la Tunisie) ont ete estimes moyennant l'approche « Data Envelopment Analysis (DEA) ». Dans un deuxieme temps, les scores d'efficacite obtenus ont ete regresses sur des variables explicatives par le biais d'un modele Tobit afin d'identifier ses determinants. Les resultats du modele DEA indiquent que les moyennes de l'efficacite technique, allocative et economique, sous rendements d'echelle constant (CRS), sont 70.7%, 85.1% et 59.7%, respectivement. Ces resultats montrent l'existence d'un manque a gagner sous forme de reduction du cout de production des exploitations de l'ordre de 39.3% en eliminant les differentes formes d'inefficacite. Les niveaux de l'inefficacite technique pure, d'echelle et allocative representent respectivement 32,3 %, 34,1% et 33,6 % du potentiel de reduction du cout. Les resultats empiriques du modele Tobit montrent un effet positif du choix varietal, du nombre de sources d'irrigation, d'appartenance au Groupement de Developpement agricole, de l'irrigation d'appoint et de la taille de l'exploitation sur l'efficacite economique. Ceci suggere l'existence d'un potentiel d'amelioration de l'efficacite moyennant l'intensification des services de vulgarisation, la lutte contre le morcellement et le renforcement des groupements de developpement agricole. EnglishThe objectives of this study are twofold. Firstly, we aim to measure the technical, scale and economic efficiencies for a sample of 170 cereal farmers in Chebika region (Central Tunisia) using Data Envelopment Analysis (DEA) method. Secondly, computed efficiency scores are regressed on explanatory variables using a Tobit model, in order to identify its determinants. Results of the DEA model show that average technical, allocative and economic efficiencies under constant returns to scale (CRS) are 70.7%, 85.1% and 59.7%, respectively. By operating at full economic efficiency levels, the sampled farms would be able to reduce their costs of wheat production with around 39.3%. Pure technical efficiency, scale and allocative efficiency levels account respectively for 32.3 %, 34.1% and 33.6 % of the total cost reductions. Results of the Tobit regression indicate the positive effect, on economic efficiency, of variety choices, source of irrigation, membership in water users association, irrigation management and farm size. This suggests that there is potential to improve production efficiency through developing and implementing targeted programs for inefficient farmers, intensification of specific and targeted extension services, and encouragement of farmers' membership to association.