Assessing Aggregate Structural Change

The aim of this paper is to investigate the evolution of Italian aggregate structural change over the years 1965–85. We use annual input–output (IO) tables in current and constant prices to derive an aggregate index of structural change. We adopt several techniques for such an explorative analysis. First, we borrow from qualitative matrix analysis and graph theory some basic concepts to assess direct and indirect links among sectors, and interrelatedness measures are derived in a straightforward way. However, qualitative analysis of indirect links may be flawed, since it can establish a path that is quantitatively negligible. Then, we turn our attention to a non-standard quantitative index derived from structural path analysis, expressed by a simple function of the input matrix determinant. Since empirical findings indicate a structural break in 1975, we derive another measure of technical change: the dominant eigenvalue. Such an index has several interesting properties but no clear relationship to the circularity process implicit in the Leontief model. Results for constant- and actual-price IO tables are discussed and compared with main macro-economic variables over the sample. Empirical findings indicate a relationship between investment, variability in final demand and aggregate structural change.