Agricultural volume index numbers and the choice of aggregation formula: an empirical investagation for the member states of the EC

Abstract

Analysis of volume index numbers of agricultural outputs and inputs has occupied agricultural economists and others for generations. In recent years these indices have become key indicators in the context of the annual policy reviews associated with the CAP. Despite the centrality of these measures in the policy formulation process, the methods employed in their compilation are rarely subject to critique, at least beyond the portals of official statistical agencies. Yet the computational procedures adopted by the SOEC and the Member States in the construction of such indices are open to many theoretical objections. One of the primary criticisms and the subject of this paper is the theoretical deficiencies in the aggregation formulae used in the construction of the index numbers. More pertinent, however, in the author's judgement is the empirical relevance of the theoretical shortcomings. In other words, do the various possible aggregation procedures produce broadly similar findings? If the answer is in the affirmative then the cost of employing theoretically 'soft' methodologies is thereby diminished. For instance, one might be prepared to abide by a theoretically deficient aggregation formula, in the context of making inter-country comparisons, if the ranking of countries were unaffected. If the reverse is true, existing practices may be worse than useless to the extent that they lead to inappropriate policy decisions. In a policy context therefore, the importance of ascertaining the results of employing naive aggregation procedures, when more theoretically favoured approaches are available is manifest.