Assessing the impacts of EU's common agricultural policy on regional convergence: sub-national evidence from Germany

Abstract

This article assesses the relationship between transfers arising from EU's Common Agricultural Policy (CAP) and convergence in both farmers’ revenues and interpersonal income redistribution using a sample of 26 regions in the state of Hesse, Germany, over the period 1979 to 2004 and 1991 to 2004, respectively. We thereby combine the concept of sigma convergence with Shorrock's inequality decomposition in order to determine the driving forces in distributional dynamics of farmers’ revenues. Additionally, we apply alternative methodologies to investigate how per capita incomes have evolved over time. Explicitly comparing the situations with and without transfers, our results indicate that the CAP tends to smooth differences in farmers’ revenues across regions, but does not impede a strong divergence through time. The latter is mainly driven by increasing structural differences between the regions, while disparities in intensity turn out to be less important. The empirical analysis also shows that CAP transfers reduce income inequality within society as a whole. However, this impact proved to be negligible in explaining distributional dynamics and growth of per capita incomes.

Acknowledgements

This article was part of a project of the Collaborative Research Centre 299, ‘Land use options for peripheral regions’. Financial support by Deutsche Forschungsgemeinschaft is gratefully acknowledged. Helpful suggestions on an earlier draft were provided by our colleagues at the Institute of Agricultural Policy and Market Research at the University of Giessen and the participants of the Spatial Econometrics Conference 2007 in Cambridge. We also wish to thank the anonymous referee for insightful and constructive comments. All remaining errors and omissions are ours.

Notes

¹ Within EU's new financial perspective ranging from 2007 to 2013 expenditures on cohesion policy will for the first time slightly exceed those for agriculture (European Union, 2005; Leonardi, 2006).

² The Nomenclature of Territorial Units for Statistics (NUTS) was established by EUROSTAT in order to provide a single uniform breakdown of territorial units for the production of regional statistics for the EU. The NUTS is a hierarchical classification, whereby NUTS-0 corresponds to the Member States and increasing numbers refer to more disaggregated levels.

³ Direct area payments for arable crops were based on the reference, i.e. the average yield of a country or large subareas within it. Regions at the more localized level were thus overcompensated if their yields are below the reference yield and vice versa.

⁴ The included commodities are wheat, rye, barley, oat, rapeseed, sugar beets, potatoes, milk, beef and veal, pigmeat and sheepmeat. For the first five, the computation of regional PSEs accounts for direct area payments. PSEs for the remaining six commodities are regionalized by using the PSE per unit (cf. first term on the right-hand side of Equation 2).

⁵ Indeed, there is only one region, which has positive net transfers of about 60 euros per capita.

⁶ This does not mean that the concept of sigma convergence outclasses beta convergence. For example, Sala-i-Martin (1995) argues for using both concepts complementarily as the former describes the distributional evolution through time while the latter investigates the mobility within that distribution.

⁷ The decrease in CAP transfers (i.e. losses) is again mainly caused by the significant policy reforms in the 1990s. Shifting to some extent from market price support to direct payments thereby lead to less negative welfare effects for society as a whole (for a theoretical analysis and discussion, see Helmberger (1991)).

⁸ Following the OECD scheme, a NUTS-3 region is defined as predominantly urban, if less than 15% of the population lives in rural local administrative units. If this ratio is between 15% and 50%, the region is defined as intermediate, and if it is more than 50%, the region is defined as predominantly rural.