A Mixed Geographically Weighted Approach to
Decoupling and Rural Development in the EU-15
Francesco Pecci*, Maria Sassi**
* Dipartimento di Economie, Società ed Istituzioni and SPERA – University of Verona – Italy – e-mail:
** Dipartimento di Ricerche Aziendali – University of Pavia – Italy – e-mail:
From a policy point of view, the CAP reform has emphasised the need for assessing:
- the impact of decoupling on the agricultural sector considering the interaction between the direct payments and the socio-economic variables sensitive to the measure, particularly those relevant for cross-compliance, modulation and rural development interventions;
- the different impact (direction and intensity) of the socio-economic variables on agriculture, in face of the importance assumed by territorial interventions in order to promote local factors of agricultural development.
In this context, the understanding of the existence of different regional models that influence the agricultural productivity has become a key issue. The literature generally faces the topic without considering the spatial patterns, associations and heterogeneity that might affect, at time also strongly, these relationships at the regional level. This is the object of the paper that, after the indicators selection according to key component of the CAP and Rural Development Policy reform and for a sample of regions at NUTS2 and 1 of the UE-15:
- identifies, by a Mixed Geographically Weighted Regression (MGWR) approach, the factors that influence the agricultural productivity and the intensity of this impact at the local level; and
- highlights, through a cluster analysis, the existence of groups of regions within which the level of agricultural productivity is influenced by homogeneous values of the non-stationary parameters.
The MGWR represents an advancement of the Geographically Weighted Regression (GWR) technique, that is a useful method to investigate spatial non-stationarity that can occur for: i) the existence of intrinsic differences in relationships over space; and ii) the inclusion into the regression equation of incorrect functional forms of relationships between variables (such a non-linear relationship between two variables being described by a linear one) and/or the exclusion of relevant variables.
Although GWR is now an established technique with increasing numbers of applications, almost all the models have been calibrated by allowing all of the relationships in the model to vary spatially, even in the cases in which it is difficult to justify why some relationships should be allowed to vary spatially. As a consequence, the empirical results may suggest that some relationships are stationary over space while others vary significantly. In these instances, MGWR models, where some relationships are allowed to vary spatially while others are held constant, would seem to be more appropriate.
Thus, the approach adopted allows for a more accurate understanding of the socio-economic factors that more than other variables affect the agricultural productivity at the local level providing useful information for decision-makers.