The invisible hand as an emergent property: a gradient flow approach
Giorgio Fabbri, Davide Fiaschi, Cristiano Ricci

TL;DR
This paper models the economy as a gradient flow in a Wasserstein space, showing how decentralized firm reallocation leads to efficient long-run equilibrium with emergent properties, supported by empirical analysis of EU firms.
Contribution
It introduces a novel gradient flow framework for general equilibrium dynamics, linking decentralized decisions to global optimization and efficiency, with extensions and empirical validation.
Findings
Long-run equilibrium maximizes aggregate consumption.
Decentralized decisions can be viewed as global optimization.
Empirical evidence of sectoral profit convergence and limited reallocation costs.
Abstract
We develop a general equilibrium model in which, at each instant, a short-run competitive equilibrium arises. Heterogeneity in factor allocation generates differential profit rates across sectors, prompting firms to move between them under myopic profit-seeking behaviour, subject to quadratic reallocation costs. The aggregate dynamics of the economy can be formalised as a gradient flow in a Wasserstein space, starting from a partial differential equation that describes the reallocation of firms across sectors. Two key emergent properties arise: (i) decentralised and uncoordinated decisions can be reinterpreted as the solution to a sequence of global optimisation problems, involving a function of aggregate consumption, which increases monotonically along the dynamic path; (ii) the long-run competitive equilibrium is efficient, as the distribution of firms maximises aggregate consumption…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsArtificial Intelligence in Games
