This paper studies how climatic cooling during the Little Ice Age may have affected the incentives to innovate in pre-industrial agriculture. The core argument is that colder regions faced stronger pressure to adapt, which increased labor effort, learning-by-doing, and the accumulation of productive knowledge. We develop a model of induced innovation and show how divergence in agricultural productivity between regions can emerge even when they face similar climatic shocks.

This paper examines how industrialization spread across Europe and emphasizes the role of spatial frictions. Using patents as a proxy for innovation, it studies how roads, waterways, coal access, and geographic connectivity shaped the diffusion of industrial activity from Britain to the rest of Europe during the 18th and 19th centuries.

With Andrea Sbarile, we extend the Hotelling framework to a two-dimensional market with heterogeneous population density. The model can generate agglomeration, dispersion, and richer equilibrium patterns than the standard linear-city setup, with the broader goal of connecting spatial competition theory more closely to realistic geographic environments. Firms offer a homogeneous good but compete in price and location simultaneously.

I am working with Marc-Antoine Faure and Federico Mij on an empirical analysis of the effects of a recent policy implemented by the Italian government on the pricing of retail gas stations. This policy, effective from 1 August 2023, requires retail gas stations to prominently display regional average oil prices alongside their own pricing.

The central inquiry of our research is to ascertain the impact of this policy on the national average price of gasoline as stated by the government. Our methodological approach involves the construction of a difference-in-differences model, which allows us to isolate the effect of the policy from other concurrent economic factors.

In this paper, I study the optimal spatial allocation of health care facilities.
I analyze the problem from the perspective of a central planner who aims to minimize the average distance between citizens and HCF, thereby achieving a Pareto-efficient allocation.
Using Italian public data, I show that the current distribution of pharmacies, largely shaped by historical and political constraints, is approximately 40% less efficient than the optimal benchmark.