Spatial Clustering with Functional Trend Data: An Application to Italian Health Tax Detractions

The integration of functional and spatial data in clustering methods is increasingly relevant in regional and urban economics. This paper introduces a novel spatial clustering algorithm that simultaneously considers geographical proximity and the similarity of temporal trends to identify territorially coherent clusters. The methodology, validated through simulations and applied to realworld data on Italian municipal health tax detractions, reveals significant geographical groupings characterized by similar levels and dynamics of tax benefits. Findings highlight that tax detractions align more closely with the economic capacity of individuals and areas rather than actual healthcare needs, raising concerns about equity and the territorial distribution of fiscal advantages. This divergence between formal universal healthcare principles and practical fiscal outcomes underscores the need for spatially aware policy tools. The proposed approach offers a replicable framework for analyzing spatio-temporal patterns across various domains, balancing interpretability and methodological rigor.