The financial sector is a critical component of any economic system, as it
delivers key qualitative asset transformation services in terms of liquidity, maturity
and volume. Although these functions could in principle be carried out separately by
specialized actors, in the end it is their systemic co-evolution that determines how
the aggregate economy performs and withstands disruptions. In this paper we argue
that a functional perspective on financial intermediation can be usefully employed to
investigate thefunctioning of financial networks. We do thisintwosteps.First,we use
previouslyunreleaseddatatoshowthatfocusingontheeconomicfunctionsperformed
over time by the different institutions exchanging funds in an interbank market can
be informative, even if the underlying topological structure of their relations remains
constant. Second, a set of alternative artificial histories are generated and stress-tested
by using real data as a calibration base, with the aim of performing counterfactual
welfare comparisons among different topological structures.