Essays on economic modelling of farmers’ behaviours : special cases of structural change, technology adoption and policy changes

Abstract: Multiple socio-economic and biophysical factors affect farmers’ behavioural responses to economic policy and technology adoption, but personal beliefs and expert opinions are also important for making actual choices on farm policy and agricultural innovations. Using econometric and mathematical programming models, this thesis assessed farmers’ behavioural responses to farm structural changes, technology adoption and policy interventions. For structural changes, the results showed that conversion to organic farming in the European Union (EU) is a two-tier decision, with a first choice of staying conventional or moving to conversion and a second choice of mixed or organic production. This gradual conversion process encourages the presence of nested structures within the process, largely influenced by socio-economic and biophysical variables such as milk prices, policy incentives and technology factors. Following milk quota abolition under the EU Common Agricultural Policy (CAP) reform, risk of price uncertainty emerged as another key factor for the organic conversion process of Swedish dairy farms. In econometric modelling of farmers’ choices and impacts of responses to adopting agricultural technology, the parametric econometric specification is commonly applied. However, it cannot guarantee the true specification of behavioural responses to agricultural innovation. In this thesis, a recently developed nonparametric (NP) kernel density estimator was applied in impact assessment of agricultural technology adoption, using the case of zero tillage technology in the rice-wheat cropping system of the Indo-Gangetic Plains (IGP) region. This estimator can capture possible nonlinearities in the data generation process that cannot be known a priori. The results showed that the NP specification outperformed the parametric specification in predicting propensity scores and produced impact estimates with small standard error. For the study area, the results showed that introduction of the new technology generated the economic benefits of markedly lower tillage costs yield in zero-tilled plots. In a study of policy interventions, the wealth effect of the CAP direct payment system on agricultural crop production decisions was analysed using Bayesian econometrics and positive mathematical programming (PMP). Under risk, lump-sum payments may influence risk-averse farmers on crop production decisions. In simulations, no direct payment in a risky environment caused a shift in land use away from risky crops towards low-risk crops, altering the crop mix. Moreover, the farm-level wealth effect varied greatly between farms, although its magnitude was influenced by regional characteristics, e.g. historical farm structure and region-specific conditions.