Agricultural Sustainability from a Societal View: An Analysis of Southern Spanish Citizens

Abstract

Sustainable agriculture refers to farming systems with economic, social, and environmental viability that must respond to citizens’ interests and concerns. However, European citizens are not satisfied with the Common Agricultural Policy (CAP) due to misinterpretation of their preferences. Because of this, the European agricultural model’s long-term viability is being questioned, especially after the European Commission’s CAP proposals in 2011. This paper examines European agriculture’s potential sustainability with regard to citizens’ preferences. First, focus groups and the Analytic Hierarchy Process are used to identify and quantify southern Spanish citizens’ preferences for farming. Second, socio-demographic features and opinions that determine preferences towards agriculture are studied by a multinomial logit model and a cluster analysis. A comparison is made between citizens’ preferences and the CAP aims because the CAP aims address all European farming. The main results indicate that agricultural economic, environmental, and social functions are equally important to the respondents in our study, even though the CAP prioritizes the economic ones. However, some citizen groups agree with the agricultural model designed by the CAP.

Appendices

Appendices

Appendix 1: AHP Application

When the pairwise comparisons were conducted, weights (vector of priorities) were estimated from Saaty’s matrices (A = a_ijk), formed by the reciprocal of the paired comparisons of each element.

Various methods have been proposed to estimate the weights. Saaty proposed the eigenvector method as the best estimator; however, the literature does not provide any evidence for the superiority of any method (Fichtner 1986; Zahedi 1986). We opted for the row geometric mean to apply the re-sampling bootstrap method² to determine the statistically significant differences among the estimated weights. Then the geometric mean was applied to aggregate the individual pairwise comparisons $\left( {{\text{a}}_{\text{ij}} = \root{\text{n}} \of {{\prod\nolimits_{{{\text{i}} = 1}}^{{{\text{i}} = {\text{n}}}} {{\text{a}}_{\text{ijk}} } }}} \right)$ and establish the weights for the entire group using the so-called aggregation of individual preferences (AIP). Aczél and Saaty (1983), Golden et al. (1989), and Forman and Peniwati (1998) suggested the geometric mean as the most suitable method of aggregation.

Appendix 2: Multinomial Logit Model: Explanatory Variables

The explanatory variables introduced in the Generic demand multinomial logit model were age (linear); place of residence (rural, urban and metropolitan); farmers in the family (without, first degree and second degree); preferences for taxes on trade or non-trade functions (trade, non-trade and both); opinion of the CAP’s impact on agriculture (negative, neutral and positive); opinion of agricultural importance for the future of Andalusia (linear); extent to which the respondent agreed with not continuing to support agriculture unless it protects the environment and creates jobs (disagree, medium and agree); Factor_Environment; Factor_Production; and Factor_Social.

The last three variables, the factor variables, represent Andalusian citizen’s views on how agriculture fits into the eleven Specific demands (evaluated by a Likert scale with 1 being very low and 5 being very high). To reduce the data, a factor analysis was performed and three factors were detected (Table 7).

Table 7 Factorial analysis: degree to which agricultural fulfils specific demands